ill! ,tfie<-'- p or CALIFORNIA. ?AMmjCNT Or CIVIL ENCMNEI BERKELEY, CAU w. PLATE I. OPTICS or THE COMPOUND MICROSCOPE. 1 FI Upper focal plane of objective. F 2 Lower focal plane of eyepiece. A Optical tube length = distance between FI and F 2 . Oi Object. O 2 Real image in F 2 transposed by collective lens. 3 Real image in eyepiece diaphragm. 04 Virtual image formed at the projection distance C, 250 mm. from EP. EP Eye-point. CD Condenser diaphragm. L Mechanical tube length (160 mm.). i, 2, 3 Three pencils of parallel light coming from different points of a distant illuminant. 1 From " The Microscopy of Drinking Water " by George C. Whipple. Reproduced through the courtesy of the author and that of the Bausch & Lornb Optical Co. ELEMENTARY CHEMICAL MICROSCOPY BY EMILE MONNIN CHAMOT, B. S., PH.D. \ Professor of Sanitary Chemistry and Toxicology, Cornell University FIRST EDITION FIRST THOUSAND NEW YORK JOHN WILEY & SONS, INC. LONDON: CHAPMAN & HALL, LIMITED 1915 CJT Engineering COPYRIGHT, 1915, BY EMILE MONNIN CHAMOT Stanhope ipress F. H. GILSON COMPANT BOSTON, U.S.A. PREFACE. The American chemist, usually ready to accept with alac- rity all time, labor and money saving devices, has been strangely backward in taking advantage of the benefits to be gained through the intelligent application of chemical microscopic meth- ods in the industries and in research. He has also failed to grasp the fact that the modern microscope is, in reality, a more important adjunct to his laboratory than spectrometer, polarim- eter or ref ractometer ; in fact, it may be said that the micro- scope is entitled to as important a place as the analytical balance. No one other instrument can perform so many functions and do them all well. This curious reluctance to grasp the opportunities offered is the more extraordinary, when we recall that the earliest com- prehensive work dealing with microchemical methods was from the pen of an American Theodore G. Wormley whose classic "The Microchemistry of Poisons" appeared in 1867. The failure of the chemists to obtain from the microscope all that the instrument is capable of yielding is, perhaps, largely due, first, to the fact that few of them are given an opportunity of becoming sufficiently familiar with the instrument and its accessories; second, they are not aware of the great variety of problems which are solvable through the microscope, nor of the specific sort of problems for the investigation of which this is the instrument par excellence; third, there has been a lack of elementary manuals covering the field, and for this reason the microscope has been looked upon as an instrument peculiar to the biological laboratory. One application, if no other, should appeal to every chemist, that of microscopic qualitative analysis, because of its enormous saving of time, labor, material and space, yet with increased sensitiveness of tests and greater certainty of results. iv PREFACE The very apparent need of including a course in the manip- ulation and applications of the microscope in the curriculum of students of chemistry led to the establishment, by the author, of laboratory courses in chemical microscopy some fifteen years ago. These courses have comprised informal lectures, demonstra- tions and laboratory practices. The students have been guided by their notes and by mimeographed and typewritten sheets. With the growth of the courses in number of students, apparatus and laboratory equipment, some more permanent and compre- hensive outline has become imperative. The result has been the preparation of the present little book. The author has intended it primarily for his students in elementary chemical microscopy and as a basis for more advanced work in specific fields, but he hopes that the gathering together of methods and apparatus may prove of value to American chemists at large and perhaps serve to arouse in some an interest in one of the most fascinating branches of chemical science. The actual nucleus about which the various parts of the book have grown is a series of some twenty articles written by the author between the years 1899 and 1902 for the Journal of Applied Microscopy, dealing with methods of microchemical analysis; to this foundation have been added the laboratory direction sheets and the substance of the lectures delivered. Until the year 1911, when Emich's excellent little Lehrbuch der Mikrochemie appeared, there was not in existence any work embodying the broad applications of the microscope to the solving of problems such as arise in the chemical laboratory. So far as the writer is aware this is the only book touching this field. The topics presented by Emich are substantially those which have been covered in the author's courses with the excep- tion that more weight is placed upon analytical methods and less upon apparatus. The present writer therefore feels that there is still room for an outline of Chemical Microscopy proper. It is assumed that the students for whom this textbook is intended have had a course in crystallography and one in physics, including optics. Therefore, only a mere statement of funda- mental facts has been thought essential, that is, only so much PREFACE V as is necessary to recall knowledge already acquired but not yet applied in practice. In discussing the polarizing microscope, only the barest pos- sible outline of its use and application has been thought wise. This chapter is intended to be largely suggestive in character and to induce at least some students to extend their studies to include optical crystallography and petrography. In the chapter dealing with grinding, polishing and etching, it was found impossible to properly present the subject without unduly enlarging the book and encroaching too deeply into the field of microscopic metallurgy; only the most fundamental methods of alloy treatment have therefore been given. The instruments figured (and the methods described) have all been tested and tried by the author with but one or two excep- tions. The instruments are those with which the Cornell Uni- versity Laboratories are supplied or those which have kindly been loaned by their makers. Doubtless there are other pieces of apparatus and other instruments which may be as satisfactory, but it has been thought best to discuss only such as have actually been examined and tested experimentally by the author and his students. For the benefit of those who may wish to obtain similar in- struments the manufacturers have in most cases been indicated. In preparing such an outline the work of an author must of necessity be largely one of compilation, of modification of old methods and the presentation of old ideas from a new viewpoint. The present writer, therefore, makes no claims for originality, and as a student of that remarkable teacher, the late Professor Behrens of the Polytechnic School of Delft, he naturally has followed and favored the methods developed by this master of the art of the qualitative analysis of minute quantities of material and he acknowledges fully his indebtedness to his former teacher, and takes this opportunity of expressing his gratitude for the advice and help given him by his guide and friend. To Simon Henry Gage, Professor Emeritus of Histology and Embryology, the writer also acknowledges his indebtedness for much that is here presented. It is largely due to the spirit of VI PREFACE optimism and love for research with which this indefatigable in- vestigator is ever surrounded that the author was originally led to enter the field of applied microscopy when first a student. To Professor Louis Munroe Dennis, Head of the Department of Chemistry of Cornell University, the writer is even more in- debted in later years for his unflagging enthusiasm and confidence in the possibilities of a neglected field. Without his encourage- ment and support, the development of laboratories and equipment would have been impossible and the preparation of this little book impracticable. The author also wishes to express his indebtedness to Dr. E. Mace of the University of Nancy, France, and to colleagues in the Cornell University departments of chemistry, physics, and mineralogy for valuable advice and suggestions. His thanks are also due to his assistants Dr. C. M. Sherwood and Mr. H. I. Cole for reading manuscript and testing methods. E. M. C. ITHACA, N. Y., June, 1914. CONTENTS. Optics of the Compound Microscope Frontispiece CHAPTER I. OBJECTIVES AND OCULARS. PAGE Function of the objective i Designation of objectives i Working distance of objectives 2 Different kinds of objectives 2 The draw-tube 4 Angular aperture of objectives 4 Numerical aperture of objectives 5 Immersion objectives 5 Variable objectives 6 Resolving power 6 Illuminating power 8 Penetrating power -8 Selecting objectives 8 Care of objectives 9 Function of oculars 1 1 Negative and positive oculars n Eye-point 12 Different types of oculars 13 Care of oculars 14 Limit of magnification 15 CHAPTER II. MICROSCOPES FOR USE IN CHEMICAL LABORATORIES. Specifications for chemical microscopes 17 Microscopes for general chemical microscopy 19 Large stage microscopes 22 Comparison oculars 23 Comparison microscopes 25 Hot stage microscopes 28 vii viii CONTENTS CHAPTER III. ILLUMINATION or OBJECTS; ILLUMINATING DEVICES. PAGE Different modes of illumination 30 Transmitted light 30 Condensers, Abbe condensers 32 Reflected light 37 Dark-ground illumination 40 Dark-ground illuminators 41 Objectives for use with dark-ground illuminators 43 Resolving power with dark-ground illuminators 44 Adjustment of dark-ground illuminators 46 Orthogonal illumination 50 Differential color illumination 51 Ultraviolet ray illumination 51 Fluorescence microscope 52 CHAPTER IV. ULTRAMICROSCOPES; APPARATUS FOR THE STUDY OF ULTRAMICROSCOPIC PARTICLES. The principle of the ultramicroscope 54 Brownian motion 55 The diffraction images of ultramicroscopic particles 55 The slit ultramicroscope and its adjustment 57 Reflecting condenser ultramicroscopes 65 The cardioid ultramicroscope 67 The reflecting prism ultramicroscope of Cotton and Mouton 69 The Jentzsch reflecting condenser 71 The immersion ultramicroscope 72 CHAPTER V. THE EXAMINATION OF OPAQUE OBJECTS; VERTICAL ILLUMINATORS; METALLURGICAL MICROSCOPES. Simple vertical illuminators 76 Adjustment of vertical illuminators 78 Interpretation of appearances 80 Special forms of vertical illuminators 81 Maintaining the alignment of illuminator and radiant 87 Mounting polished specimens for study 89 Metallurgical microscopes, metallographs 90 Shop or Works microscopes 99 CONTENTS ix CHAPTER VI. USEFUL MICROSCOPE ACCESSORIES; LABORATORY EQUIPMENT; WORK TABLES; RADIANTS. PAGE Drawing cameras 102 Drawing eyepieces 105 Microspectroscopes 106 Calibration of microspectroscopes no Mechanical stages 113 Rotating and orientating devices 115 Lens holders 118 Reagent containers 119 Rods, platinum wires, pipettes 121 Spatulas 122 Forceps 122 Object slides 123 Watch glasses and evaporators 126 Burners for microchemistry investigations 126 Tongs 129 Work tables 129 Radiants 132 Nosepieces and objective changers 137 Sedimentation glasses 139 The microscope as a polarimeter 139 CHAPTER VII. MICROMETRY; MICROMETRIC MICROSCOPES. Different methods of measuring microscopical objects 142 Methods of direct comparison 143 Micrometric microscopes 144 Micrometry with a camera lucida 147 Determining the magnifying power of a microscope 148 Micrometry by means of micrometer oculars 149 Determination of the ocular micrometer ratio 150 Step micrometers 153 Contrast micrometers 154 Filar micrometers 154 Micrometry by means of a scale projected by the Abbe condenser 155 Use of the micrometer fine adjustment 157 Measurements of thickness 158 CONTENTS CHAPTER VIII. POLARIZED LIGHT; THE SIMPLE POLARIZING MICROSCOPE; CRYSTALS UNDER THE MICROSCOPE. PAGE Polarized light, its properties and applications 159 Isotropic and anisotropic substances 159 The Nicol prism 160 Polarizers and analyzers 162 Testing the adjustments of the polarizing microscope 163 Centering the rotating stage 164 Polarized light without a Nicol prism 166 Fundamental principles of crystallography 167 Elements of optical crystallography 170 Directions or axes of vibration of crystals 172 Use of converging polarized light 173 Polarization colors 174 The selenite plate 175 Absorption of light, pleochroism 176 Measurement of angles and of extinction angles 177 Characteristics of the six crystal systems 180 Experiments with crystals 182 CHAPTER IX. THE DETERMINATION OF REFRACTIVE INDEX BY MEANS OF THE MICROSCOPE. The relation between refractive index and contour bands 184 Principle of the immersion method 185 Behavior of air bubbles and oil globules 186 Half-shadow method of illumination 188 Calculating refractive indices of mixtures of liquids 191 Refractive index of anisotropic crystals 193 Uniaxial and biaxial crystals 193 Determination of the refractive index of liquids 196 Determination of thickness by refractive index 200 Liquids for use in the immersion method 201 Crystals for use in the immersion method 203 Refractive indices of typical crystals .' 204 CHAPTER X. QUANTITATIVE ANALYSIS BY MEANS OF THE MICROSCOPE. Methods available 205 Analysis of powdered material 206 Net ruled oculars and their uses . . 208 CONTENTS xi PAGE Counting cells 210 Determination of weight by micrometry 212 Volume and weight per cents; area measurements 216 Estimation of molecular weight 216 CHAPTER XI. THE DETERMINATION or MELTING AND SUBLIMING POINTS. Approximate methods 220 Exact methods 222 Hot stages 224 Subliming points 227 CHAPTER XII. METHODS FOR HANDLING SMALL AMOUNTS or MATERIAL. Testing for solubility 228 Decantation 230 The centrifuge 233 Filtration 238 Sublimation 240 Distillation 244 Ignition, fusion 248 Grinding and mixing 249 CHAPTER XIII. THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS. The various ways in which reagents are applied 250 Preparation of special reagents 271 CHAPTER XIV. CHARACTERISTIC MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS AND ACIDS WHEN IN SIMPLE MIXTURES. CATIONS: Sodium 273 Potassium 281 Ammonium 285 Calcium 287 Strontium 292 Barium 295 Xll CONTENTS PAGE Calcium, strontium and barium, additional tests 300 Magnesium 304 Zinc 307 Cadmium 316 Mercury 318 Lead 323 Silver 330 Copper.. 339 Aluminum 341 Tin 346 Arsenic 348 Antimony 351 Bismuth 354 Chromium 356 Manganese 359 Iron 362 Nickel 363 Cobalt 365 Testing for cations in simple salts 367 ANIONS: Testing for anions in simple salts 368 Group reactions of the anions 369 Acetates 374 Arsenates 374 Arsenites 374 Borates 375 Bromides 375 Carbonates 375 Chlorides 376 Chlorates 376 Chromates, bichromates 376 Cyanides 376 Cyanates 377 Ferricyanides 377 Ferrocyanides 378 lodates 378 Iodides 378 Nitrates 378 Nitrites 379 Oxalates 379 Phosphates 380 Silicates 380 Sulphates 380 Sulphites, thiosulphates 381 Sulphides 381 Sulphocyanates 381 Tartrates. . 381 CONTENTS xiii CHAPTER XV. PREPARING OPAQUE OBJECTS FOR THE MICROSCOPIC STUDY OF INTERNAL STRUCTURE. PAGE Fundamental principles 383 Grinding; abrasive wheels 384 Grade and grain of abrasive wheels 385 Selecting wheels 386 Speed of rotating wheels 387 Abrasive papers 388 Preparing specimens 389 Etching 391 Etching liquids 392 APPENDIX. Table of melting points 395 Periodic classification of the elements 396 Books for reference 397 INDEX 399 ELEMENTARY CHEMICAL MICROSCOPY. CHAPTER I. OBJECTIVES AND OCULARS. The modern compound microscope, in any one of its many complicated forms employed by chemists, consists essentially of three parts, (i) an objective, (2) an eyepiece or ocular and (3) a device for properly illuminating the object. The manner in which these three essential components are mechanically mounted, and their relative importance with respect to each other will de- pend upon the nature of the investigation to which the instru- ment is to be specifically applied. The mechanical parts of the microscope can therefore be best discussed under the different types of microscopes applied 'to special investigations. The optical components, however, need a few words in order that the student may refresh his memory relative to the optics involved. Objectives have as their function the formation of an enlarged real image of the object placed upon the stage of the microscope. From the viewpoint of the chemist, their construction should be such as to keep them as far above the object as possible, yet yield an image of as great an area of the object as can be ob- tained without distortion and without color bands or fringes. In addition, they should possess considerable depth of focus. Objectives are commonly designated by their equivalent focal length, as, for example, i inch, 32 millimeters, etc., the numbers indicating that the objective will produce a real image of approxi- mately the same size as that produced by a simple convex lens 2 ELEMENTARY CHEMICAL MICROSCOPY whose principal focus lies at the distance marked upon the ob- jective. In a similarly constructed series, the smaller the value of the equivalent focus, the greater will be the magnifying power of the objective. A few manufacturers still arbitrarily letter or number their objectives. In such cases it is generally the rule that the earlier in the alphabet the letter or the smaller the number in the series the lower the magnifying power. When properly focused upon a preparation, the front or lowest lens entering into the construction of an objective is usually nearer to the preparation (in dry objectives) than the distance indicated by the equivalent focus. This distance between the front combination of the objective and the preparation, when in focus, is known as the working distance of an objective. In their selection for use in microchemical analysis the working distance becomes one of the most important considerations affecting the choice of the objectives. The construction of typical microscope objectives is shown diagrammatically in Figs. 17, 1 8 and 19, page 44. All objectives are corrected to a greater or lesser degree for chromatic aberration (presence of colored fringes around the images) and also largely for spherical aberration (failure to yield a flat field of view) . When the spherical aberration is so corrected as to yield an especially large and flat field the objectives are often called aplanatic objectives. Although an objective may be so corrected as to yield a flat field, images of objects lying near the circumference are apt to be hazy or indistinct, the result of a form of spherical aberration known as coma; this is especially marked in high power objectives and requires unusual care in construction for its elimination. 1 In all ordinary so-called achromatic objectives the corrections are usually such as to bring the rays of two spectral colors to a focus. In such lenses the optical and chemical foci may lie in different planes and therefore such objectives may not give really good results if employed in photomicrography; for this reason specially corrected achromatic lenses called photo- 1 See Spitta, Microscopy, London, 1909. OBJECTIVES AND OCULARS 3 objectives are manufactured. When in the correction for chro- matic aberration three spectral color rays are brought to a common focus the objectives are known as apochromatic objectives. In these objectives the chemical and optical foci are identical and we have the highest grade of lenses at present available. Al- though in apochromatic objectives rays of three colors are brought to a correct focus, the images produced by these three sets of rays are not coincident and thus yield a colored fringe or halo at the edges of the field. This, however, is eliminated by em- ploying slightly over-corrected eyepieces, known as compensating eyepieces, in which the construction is such as to neutralize, or compensate for, the errors due to the objectives. Beautifully clear, colorless images are thus obtained, but the field is rarely flat. Objectives are either dry or immersion according as they are designed to be used with air or with some liquid between the front or lower lens and the preparation. High power dry ob- jectives must each be specially adjusted for a certain definite thickness of cover glass. In order to permit some freedom of choice in cover glasses most high grade high power dry objectives are adjustable and are provided with a movable graduated collar, permitting the regulation of the objective for the thickness of the cover glass used; that is, a part of the combination of lenses making up the objective may be raised or lowered in the mount- ing, thus affording a correction for the displacement of the image brought about by the cover glass. By consulting the diagram, Fig. 19, page 44, it will be seen that by turning the collar C the combination of lenses L will be displaced and their distance from the combination L' will either be increased or diminished. A cover glass which is thicker than that for which the objective is corrected affects the image in the same manner as if the spheri- cal aberration were over-corrected, while on the other hand if too thin the effect produced is similar to that of under-correc- tion. In the first case the focal distance of the objective must be increased, and in the second, decreased. This is accomplished by turning the adjusting collar to the right or left, as the case may require, or, in the absence of such a device, by shortening or lengthening the distance between the eyepiece and the objec- 4 ELEMENTARY CHEMICAL MICROSCOPY tive, shortening for cover glasses too thick, and lengthening for those which are too thin. Fitting into the body tube of modern microscopes is a tube which may be drawn out several centi- meters. This tube is known as the draw-tube and is graduated in millimeters. Objectives are commonly corrected (for use on the usual type of microscope) for a tube length of 160 milli- meters. The i6o-millimeter mark will therefore be found only when the draw-tube is pulled out a short distance. This position of the standard mark permits lengthening or shortening the draw- tube, and thus correcting for cover glass thickness as stated above. In addition to corrections for chromatic and spherical aberra- tion at least two other factors must be taken into account in comparing, or choosing between, objectives of similar equivalent focal length. These are the angular aperture and the numerical aperture of the objectives. By the angular aperture of an ob- jective is meant the " angle contained, in each case, between the most diverging rays issuing from the axial point of an object {i.e., a point in the object situated on the optic axis of the micro- scope), that can enter the objective and take part in the formation of an image" (Carpenter-Gage). This angle is obviously that of the cone of light rays whose -apex lies in the optic axis of the microscope at the point where the axis passes through the plane of the object and the diameter of whose base is equivalent to the opening of the front lens com- bination of the objective. Dry objectives may be compared with each other with refer- ence to their angular aperture. In general the angular aperture depends largely upon the diameter of the front combination of the objective, and usually in objectives of like magnifying power, the greater this diameter the larger will be the angular aperture and the wider and clearer will be the area or field covered. It is also generally true that the shorter the equivalent focus of the objective, the larger its angular aperture and that dry objectives of small working distance usually have large angular apertures. It is obvious that in dry objectives an easy comparison of the relative areas of field covered is afforded by a consideration of .angular apertures. The true field of view of a compound mi- OBJECTIVES AND OCULARS 5 croscope is, however, controlled by the ocular, as will be seen below. It would appear at first sight that the light-grasping power of an objective is indicated by its angular aperture. Such is not the case, for Abbe has proved that in comparing objectives as to their light-grasping and transmitting power it is the sine of half the angle of aperture which should be taken into account and not the angular aperture; and further, that since objectives are not all dry, the index of refraction of the medium between the objective and the object must necessarily be considered. It is therefore now conceded that the light-grasping and transmitting power of an objective is equal to the refractive index of the medium in which the objective dips multiplied by the sine of half the angle of aperture. The product is what is known as the Numerical Aperture and is expressed N.A. = n sin /*. If the above formula is accepted as true it is evident that if the value of n is increased the numerical aperture will likewise be increased. The light rays illuminating an object by transmission through the preparation evidently pass from a denser medium (object) to a rarer medium (air), and following the law of refraction are bent away from the perpendicular. Hence part of these light rays are lost, since they are bent so far that they cannot enter the small front lens of the objective. To prevent this loss and secure a brilliant image it is necessary, according to the formula N.A. = n sin /z, to increase the value of n. Therefore, to obtain very high powers, the substitution of some liquid for air (n = i) between the objective and the preparation becomes imperative in order that the image may be bright and distinct. 1 Objectives permitting the use of a liquid in this manner are known as immersion objectives. When water is employed (n = 1.33) they are called water immersion, and when an oily liquid, oil immersion. Usually the oil consists of slightly thick- ened oil of cedar wood (n = 1.52), and since the refractive index of glass object slides and cover glasses is approximately 1.52 1 Abbe found that the brightness of the image varies as the square of the numer- ical aperture. ELEMENTARY CHEMICAL MICROSCOPY also, such objectives are more commonly designated homog- enous immersion objectives. Alpha mono-brom naphthalene is also sometimes used as an immersion fluid (n = 1.66) and gives us the highest numerical aperture obtainable. 1 Since oil immersion objectives have the highest numerical apertures they therefore yield the brightest and the clearest images, and repre- sent the highest development in the art of microscopic objective manufacture. In the case of immersion objectives the working distances are usually greater than the equivalent foci. Variable Objectives are so constructed that the distances be- tween two sets of component lenses may be changed by means of a graduated collar, permitting a wide range in the magnifying power of the objective. A single objective is thus made to do the same work as a number of objectives of fixed system. For low powers, the chemist will find an objective of this sort an ex- ceedingly great convenience. Fig. i shows a variable objective as manufactured by Zeiss. Its range of magnification lies be- tween 29 and 43 diameters and its free workin & distance between the limits 53 millimeters and 13 millimeters. To obtain a similar range with non- variable objectives requires four or five. Variable objectives do satisfactory work and are rela- tively inexpensive. A measure of the quality of an objective lies in its ability to make clear any fine and delicate details of structure. It is, therefore, customary to speak of the resolving power of objectives and express this attribute in terms of the number of fine lines per unit length the different objectives will render distinctly visible, 1 An Abbe condenser of the commonly purchased form has as its maximum a N.A. of i. 20; while the three lens condensers of the highest type will trans- mit rays only up to a numerical aperture of 1.40. Unless therefore a special achromatic condenser is available, it is manifestly useless to employ alpha mono- brom naphthalene immersion objectives since only a part of the full aperture will be available. OBJECTIVES AND OCULARS 7 or, in other words, the resolving power of an objective can be denned as the minimum distance apart two lines or spots may be and yet appear as two distinct individuals. The resolving power of an objective is dependent upon its light collecting and light transmitting power; this in turn is governed by the numerical aperture and by the particular wave length of light entering the lens system. From the viewpoint of the physicist the resolving power of an objective can be expressed as equivalent to , where X is the wave length of light. This is based upon the assumption that the illuminating cone of light completely fills the aperture of the objective. From this formula, we find that, theoretically, the limit of resolution will be attained when the magnification of an objective reaches about 900. The chemist is not alone interested in the brightness of image and the resolving power of an objective, but he is vitally con- cerned with another property, namely, the ability of the objective to make clear objects or structures in more than one plane. This is known as its penetrating power. The penetrating power of an objective has been shown to be inversely proportional to the numerical aperture and to vary as the square of the equiv- alent focus. Leaving out of consideration the numerical aperture, it is found that the resolving power of an objective is inversely pro- portional to the wave length of light. By employing light rays of very short wave lengths we may thus obtain exceptional resolution. In the above consideration it has been assumed that the illu- minating cone of light completely fills the aperture of the objective. Nelson 1 has shown that in practice with the older types of objective we can rarely count upon more than three-fourths of the available numerical aperture. More modern objectives perform somewhat better. In comparing objectives as to their ability to render struc- tures clear and distinct it is usual to do so by computing the 1 J. Roy. Micro. Soc., 1893, 15-17. 8 ELEMENTARY CHEMICAL MICROSCOPY number of ruled lines to the inch or millimeter each one will make clearly visible (resolve). To obtain these values the recip- rocal of the above given standard formula must be taken. Since, as pointed out, we cannot obtain the theoretical resolving power in practice a correction coefficient must be introduced into our formula. Nelson assigns to this coefficient the value 1.3. The practical working formulas then become : 1 Available resolving power = ' /N A \ 2 Available illuminating power = i - 1 ) > 1.3 X / Available penetrating power For white light a mean value may be assumed to be X = 5607 (= 0.5607/1) and for blue light X = 4861 (= 0.4861 /*). Advantage has been taken of the increased resolving power attainable by short wave lengths in the application of ultra- violet light (X 25oo) to photomicrography. In this way a resolving power of three times that obtainable with red light (X 75oo) may theoretically be obtained. Since ordinary glass is opaque to rays below X 3000, it is essential that the condenser, objectives, oculars, object slides, etc., be made of quartz. For similar reasons quartz is preferable to glass in all ultramicroscopy, moreover, most glass exhibits a marked violet fluorescence under the influence of ultraviolet rays; quartz does not. SELECTING OBJECTIVES. It is evident from the above briefly outlined considerations, that the choice of an objective of a given equivalent focus and magnification must depend upon the nature of the work the objective will be required to perform. In microchemical analy- sis, because of the rather unusual conditions which obtain, ob- jectives must be selected with special reference to long working distance and great depth of focus; the brightness of field and the resolving power necessarily lost are, in this class of work, of 1 Nelson, J. Roy. Micro. Soc., 1906, 521. OBJECTIVES AND OCULARS 9 little importance, since only low powers are employed and the indices of refraction of objects and surrounding medium are generally sufficiently different to permit an easy study of the preparations. When magnifications of from 300 to 500 are required in microchemical examinations, difficulty will be expe- rienced in obtaining suitable objectives unless the prospective purchaser stipulates long working distances, since the working distance of those manufactured for the use of biologists is far too short to permit their application to the study of uncovered and therefore thick drops of liquid. For the study of objects lying in a single plane, for polished surfaces, rulings, fine etchings, etc., in which sharpness of out- line and delicacy of structure or tracery are present, flatness of field and high numerical aperture are essential. Our choice is, consequently, here restricted to aplanatics or to apochromatics, bearing in mind the fact that the resolving power of an immersion objective, where applicable, is greater than that of a dry one. If, on the other hand, the investigation to be conducted involves much photomicrographic work, photo-objectives, apo- chromatics, or better still, the very carefully constructed micro- planars, microsummars, or microanastigmats, should be selected. For in addition to the fact that the chemical or actinic rays are not properly brought to a focus, it should be remembered that ordinary microscopic objectives are corrected for a fixed tube length, usually 160 millimeters, while in the case of photo- graphic work the distance between objective and plate holder is variable and in all cases much greater than the standard tube length. THE CARE OF OBJECTIVES. Objectives should always be most carefully handled and pro- tected from dust and vapors. They should be kept dry and clean by wiping with clean new lens paper. 1 Never use a piece of lens paper more than once, nor touch the lenses of objectives or oculars with the fingers or with cloths. 1 "Lens paper" is a soft absorbent tissue-like paper made from long flexible fibers expressly for cleaning lenses. 10 ELEMENTARY CHEMICAL MICROSCOPY When abrasives are employed (as, for example, in metallo- graphic work) even in adjoining rooms, all lenses should first be blown upon (but not breathed upon) and then dusted off with a very soft camel's hair brush before wiping with lens paper, otherwise serious scratching of the glass will sooner or later result. Dust on the back lens combination of the objective is often responsible for great loss of definition and greatly reduces the resolving power of an objective. Dust on the rear lens may easily be seen by removing the ocular, illuminating the objec- tive to its full capacity and looking into the microscope tube. Often a screen of ground glass placed in front of the microscope mirror renders the dust particles more clearly discernible. After using an immersion objective immediately wipe off the immersion fluid with lens paper, then if the fluid is oil, wipe the lens with lens paper moistened with xylene, and finally wipe dry. Never use alcohol in cleaning objectives or any part of the microscope. Never allow an objective to remain moistened with any fluid whatsoever a moment longer than absolutely necessary. When focusing a microscope upon a preparation, first turn the body tube down by means of the coarse adjustment until the objective is closer to the preparation than is indicated by the equivalent focus of the objective, watching carefully with the head to one side to see that the front lens is not forced against the slide. Look into the microscope and slowly raise the tube by the coarse adjustment until the object is almost in focus; complete the adjustment by means of the fine adjust- ment. Never focus down while looking into the instrument. Failure to observe this simple rule is apt to lead to serious loss and considerable expense. Never change from one objective to another without first making sure that the body tube has been raised sufficiently to allow the new objective to be slipped into place without injury to the preparation on the stage or to the objective. Never handle objectives or oculars or, in fact, any parts of the microscope with dirty, greasy, or wet fingers, or when the hands are so cold as to incur danger of dropping the apparatus. OBJECTIVES AND OCULARS II Never use a high power until the preparation has first been examined and centered with a low one. Remember that it is possible to see more of the object and see it better with low powers than with high ones. Invariably work with the lowest power which will clearly define the preparation. The most common fault of the beginner is to employ too high a magnification. Oculars. The function of the ocular or eyepiece of a com- pound microscope is to magnify the real inverted image of the object formed by the objective; but in addition to this the usual type of ocular employed serves as a collector of light rays and increases the brilliancy of the image and therefore of the useful area of the field of view. Eyepieces are of two types, those in which the real image is formed inside the lens system of the ocular, and those in which the real image is formed outside the ocular. The former are known as negative or Huygenian eyepieces; the latter, as positive or Ramsden eyepieces. Oculars are designated either by their equivalent focal length, by the number of times they magnify the real image formed by the objective or by arbitrary numbers or letters based upon either equivalent focus or magnification. The shorter the equivalent focal length the higher the magnification. When designated by their magnification the figures with which they are marked indicate the number of times the real image is mag- nified. In all ordinary microscopic work negative or Huygens oculars are employed, the use of positive or Ramsden oculars being restricted to micrometer eyepieces. In the case of positive oculars the entire lens system acts as a magnifier. The usual type of negative eyepiece is shown in section in Fig. 2, with the passage of the light rays diagrammatically indicated. It will be seen on consulting the diagram that the lower or field lens, as already stated, collects the light rays and reduces the size of the image formed by the objective and is thus, optically, in reality a part of the objective system; the eye lens functions as a magnifier of the image formed by the field lens. It is evident 12 ELEMENTARY CHEMICAL MICROSCOPY - Diaphragm that the position and diameter of the diaphragm in the eyepiece greatly influence the character and size of the field lens image, and are thus largely responsible for the area of the field of the mi- croscope, and consequently are very closely associated with the resolving power of the optical combination employed. The light rays leaving the eye lens are concentrated within a tiny circle, known as the eye-point, eye-circle, Ramsden disk, or Ramsden circle. The desig- nation " eye-point" has been given to this smallest bright spot of light, since it is the proper position for the pupil of the eye when looking into the microscope. If either above or below the eye-point, light rays are lost and the image is less bright and less clear. The diameter of the eye-point is dependent upon the numerical aperture of the objective and the magnifica- tion of the microscope. It will be found upon measuring the FIG. 2. Path of Light Rays in a Negative *. r ^.T. i E * }ece y diameters of the eye-circles produced by different oculars with the same objective, that they are inversely proportional to the magnification obtained and that with different objec- tives and one and the same eyepiece, the diameter of the eye- circle varies directly as the numerical aperture of the objectives. The value of the numerical aperture in any consideration of the probable performance of different objectives of the same equivalent focus has already been alluded to. We now see that there is a close relation existing between numerical aperture and the performance of the ocular; for example, of several objectives OBJECTIVES AND OCULARS 13 of approximately the same equivalent focus, but possessing different numerical apertures, that one having the highest aper- ture will permit the employment of an ocular of much higher power and thus yield a considerably greater magnification without loss of detail. If an attempt is made to increase the ocular magnification beyond a certain limit the eye-point becomes so small that the image resulting is blurred and indistinct. This fact must be borne in mind in microchemical examinations where high mag- nifications must often be brought about by using high power oculars with low power objectives of long working distance. In order that images of satisfactory distinctness and sharpness of detail may be obtained, the optical combination for work must be such as to yield an eye-point not less than one milli- meter in diameter nor greater than the diameter of the pupil of the eye of the observer. 1 The diameter of the eye-point and the position of the plane in which it lies can easily be ascer- tained by holding a piece of thin ground glass or waxed paper over the ocular, shading it with a screen or with the hand and raising or lowering it until the bright circle seen upon the glass or paper attains its minimum diameter. Oculars to be used on the chemical microscope should have the plane of the eye-circle at such a distance above the eye-lens as to permit the adjustment of drawing or other prisms to the position of maximum brightness and diameter of field. Compensating or Compensation oculars are eyepieces specially designed for use with apochromatic objectives. They are so called because of the fact that they aid in the correcting of chro- matic aberration. Oculars are said to be par-focal when they are so constructed as to permit their interchange on the microscope without dis- turbing the focus of the instrument. 2 Compensating oculars are usually par-focal. 1 Wright, F. E., The Methods of Petrographic Microscopic Research, Bui. 158, Carnegie Inst. Washington, 1911, p. 38. 2 For a consideration of the conditions to be fulfilled in their construction, see Gage, The Microscope, p. 47. Tenth ed. 14 ELEMENTARY CHEMICAL MICROSCOPY Projection Oculars, as their name implies, are. used in photog- raphy or with the projection microscope. Their purpose is the projection of a bright and clear image upon a screen whose distance from the ocular may be varied. This is accomplished by having the eye lens of the ocular movable in the mount, thus changing the distance between eye lens and ocular diaphragm. Goniometer oculars are eyepieces provided with cross-hairs and graduated circle. They are used for the measurement of crystal angles and may be substituted for a rotating graduated stage and thus permit angular measurements on any microscope whose tube they fit. The Care of Oculars. In general the suggestions made with respect to objectives on pages 9, 10 and n apply with equal force to eyepieces. To remove cross-haired oculars grasp them firmly between the fingers by the milled head and first lift them free from any slot into which a stud upon them may fit, then remove them by a screw motion. Dust on the ocular lenses may be located by raising and turning the entire ocular, then by unscrewing and turning first the field lens, then the eye lens. If both lenses are clean and the objective is clean yet the field shows specks of dirt and ap- pears blurred, the dust and dirt will be found to be on the disk carrying the cross-hairs or micrometer scale. Exceeding great care is required in cleaning cross-hairs and micrometer plates resting upon the diaphragm of the ocular and should only be undertaken by a person having patience, care and steady nerves. Use low oculars first and confine the work whenever possible to medium powers. Have recourse to high power oculars only as a last resort, since they cut down the light to such an ex- tent as to cause fatigue and eye-strain. Always look into a microscope with both eyes open. In the study of flat preparations between slides and cover glasses, the general rule is to obtain the proper magnification chiefly by means of the objective, using a low power ocular. But in the case of irregular surfaces or curved and heaped up drops of liquid, the reverse is essential and low power objectives OBJECTIVES AND OCULARS 15 (having long free working distance) and high oculars must be adopted. The latter procedure is also indicated when employ- ing dark ground illuminators "or ultra-condensers, namely, in- crease the magnification by the ocular. Limit of Magnification. A consultation of the tables of magnification given in the catalogues of the leading makers of microscopes and microscope lenses will show that with the modern compound microscope employed in the usual manner with stock achromatic objectives and Huygenian oculars, a magnification as high as 1500 to 2000 may be obtained, and that with stock apochromatics and compensating eyepieces this may still further be increased to 3000, the upper limit of listed combinations. Theoretically there is no limit to the magnification which may be obtained. But this must not be confused with resolving power which enables us to see things clearly and permits differ- entiating one part or structure from another. Great magnifica- tion avails us nothing if the image be blurred and irrecognizable. A little thought will show that there must be a limit to the resolving power practically available beyond which we cannot go. The shortest violet rays producing the effect of light upon the average normal human eye may be assumed to have a wave length of approximately X 4000 (or 0.4 /i) 1 . It has been shown that under ordinary conditions the smallest particle which will be visible as a black spot upon a light ground must have a diam- eter equal to at least half this value (Helmholtz-Abbe). More- over, a lens, owing to diffraction, yields as an image of a point, a diffraction disk and not a point. The final image may be con- sidered as consisting of a series of diffraction disks or patterns, and if the distances between bright points are such as to cause an overlapping of the resulting disks or their surrounding circles, a blurring of the image must result. Thus we are limited, in our attempt to see and study infinitely small particles, by the sensitiveness of the human eye, on the one hand, which cannot properly respond to the stimuli of very short wave lengths, and to the fact, on the other hand, that no matter how great the mag- 1 One micron, designated by the Greek letter /*, is equivalent to one-thousandth of a millimeter (o.ooi mm.). l6 ELEMENTARY CHEMICAL MICROSCOPY nification employed we cannot bring about a separation of the overlapping rings of the diffraction patterns. The result, there- fore, must be at the best a vague, blurred, uninterpretable image or merely a diffraction pattern. If, therefore, our wave theory of light is correct, the most minute particle which we may hope to render distinctly visible by our compound microscopes by transmitted light must have dimensions of at least 0.2 /*. It should not be inferred, however, that the existence of particles many times smaller cannot be indicated, for an invisible particle may yield a large diffraction pattern, a phenomenon which makes ultra-microscopic investi- gations possible; but we must bear in mind that in the case of ultra-microscopic particles we have no picture or image of their shape or structure and that we know of their existence simply through the light diffracted by them and thus have passed far beyond the range of the resolving power of our lenses. Although it is true that the limit of resolving power, 0.2 /u, has been seri- ously questioned by men of recognized authority, it may be accepted as beyond dispute that a moderately skillful micros- copist cannot hope in practical work to carry the resolving power of his instrument beyond this limit. In ordinary work a magnification of from 750 to 900 diameters is the upper limit of true usefulness in the study of details of structure. Above this point the worker must be an exceptionally keen and skillful observer in order that he may properly interpret the appearances seen in the images formed. It is best, therefore, to make it a rule to work with low mag- nifications. CHAPTER II. MICROSCOPES FOR USE IN CHEMICAL LABORATORIES. The problems which the chemist is called upon to solve where the microscope is of great value, if not actually essential, are so diverse in their nature and the materials to be examined so varied in size, outward form, structure and composition that it is safe to say that no single instrument will ever be constructed which will meet all requirements and fulfill all conditions. Before deciding upon any given style or model of instrument the in- tending purchaser should, therefore, first carefully consider the kind of work his instrument will most frequently be called upon to perform. A microscope for microchemical analysis and applicable to the ordinary problems arising in the chemical laboratory should fulfill the following requirements: 1. The stand should be substantially built so as to be easily and safely carried about. It should permit the attachment of the usually employed accessories, such as a mechanical stage, Abbe condenser, camera lucida, polarizing apparatus, etc. A hinged pillar allowing the inclination of the microscope is a valuable feature and a great convenience. In a vertical position for work the stand should be low enough to permit observations being made in comfort, without the necessity of having either specially high stools or low tables. It is desirable that the in- strument be entirely finished in black and have as few bright reflecting surfaces as possible. 2. There should be coarse adjustment by diagonal rack and pinion of as great range as possible. When the movement of the rack is short the usefulness of the microscope is greatly restricted, since low powers cannot then be used with thick ob- jects. A sensitive fine adjustment is also an essential, and if the fine adjustment is provided with micrometer screw and gradu- 17 1 8 ELEMENTARY CHEMICAL MICROSCOPY ated head, micrometric measurements of thickness are possible, and refractrometric determinations are simplified. 3. The body- tube carrying the objective and eyepiece should be of sufficient diameter to permit the microscope being used for photography, and it should be provided with an inner grad- uated draw- tube whose lower end is tapped with standard or universal thread for the attachment of very low power objectives or of amplifiers. 4. The stage should be circular, rotating and provided with centering screws with small milled heads. The circumference of the stage should be graduated in degrees and the surface covered with hard rubber. The stage must be constructed in such a manner as to be easily removed by simply loosening the centering screws, in order that thick objects may be examined, various heating devices employed, and opaque objects to be studied by means of vertical illuminators may be brought into focus on the substage without interfering with the proper ad- justment of radiant or illuminator. 5. The substage should consist of a simple ring, raised and lowered by screw or rack and pinion, and must permit of being swung to one side from under the stage. This ring carries condenser, polarizer, auxiliary stage, heating devices, etc. The ring should be tapped at one side and fitted with thumb-screw or with some sort of locking device to hold firmly in place the accessories fitting into the substage ring. The substage ring should also be provided with a slot or other contrivance for lining up the polarizer. If the microscope is to be used chiefly for observations at high temperatures, polarization by reflection is best. 6. It is essential that the microscope be fitted with attach- ments for study with polarized light including converging as well as plane. For all ordinary problems, the best system ap- pears to be rotating prisms of the type of the Nicol prism, one placed below the stage, the other above the objective. One or both of these polarizing prisms should be mounted so as to rotate and be provided with graduated circles. It will be found to be a great convenience if the construction is such that when polarizer MICROSCOPES FOR USE IN CHEMICAL LABORATORIES 19^ and analyzer are in their proper places the planes of vibration of these prisms will be crossed without the necessity of experi- mental adjustment. 7. The instrument must be provided with a mirror, plane on one side, concave on the other, of as large diameter as possible, which permits turning over from plane to concave side when the microscope is in a vertical position without the necessity of tipping the pillar. The mirror should be mounted on a swing- ing bar to provide very oblique light and it is desirable that the bar have an extension arm in order that the mirror may be swung to give oblique light above the stage. 8. At least two of the oculars (a high power and a low power) must be fitted with cross-hairs and stud fitting into a notch or slot in the upper end of the draw-tube. 9. The objectives should be of exceptionally long working distance and in combination with the eyepieces should yield a magnification of from 15 or 20 diameters to 300 or 350 diameters for ordinary work. 10. The instrument should be of as simple construction as possible and should permit the easy and inexpensive replace- ment of parts damaged through accident. TYPES OF MICROSCOPES FOR MICROCHEMICAL INVESTIGA- TIONS. Instruments for General Use. A microscope which con- forms very closely to the specifications given above is shown in its latest model in Fig. 3. This instrument has been con- structed after specifications of the author 1 to meet most of the problems arising in chemical laboratories in which a micro- scope may be employed. In this model an attempt has been made to provide as compact an instrument as possible, having an exceptionally great distance between the optic axis and the arm, thus providing sufficient manipulative space for large ob- jects, cells, etc.; the range of the body tube is also sufficient to permit even very low powers to be used with vertical illumina- 1 Chamot, J. Applied Micros., 2 (1899) 502. Manufactured by the Bausch & Lomb Optical Co., Rochester, N. Y. 2O ELEMENTARY CHEMICAL MICROSCOPY MH FIG. 3. Simple Polarizing Microscope for Chemical Microscopy. MICROSCOPES FOR USE IN CHEMICAL LABORATORIES 21 tors, while the range of the substage screw is long enough to permit focusing the substage ring with auxiliary stage attached in metallographic work, thus keeping the body tube with an illuminator in line with the radiant. The milled heads of the stage centering screws have been made much smaller and shorter than usual in order that they may interfere less with manipulations on the stage and be less subject to displacement. The revolving stage with circle graduated into degrees is removable by merely unscrewing the centering screws, and then lifting out the stage. This permits inserting into the substage ring an auxiliary stage for use with thick objects, or opaque objects, to be studied with a vertical illuminator (see Fig. 41, page 88), or when preparations are to be heated with a tiny flame. The polarizer PO consists of a Nicol prism set in a rotating mounting graduated into degrees. A stud in the fixed part of the mounting fits into a slot in the substage ring, thus insuring that the polarizer mounting is always in the same relative position. The analyzer, PA, a Thompson prism, fits over the eyepiece, rotates, and is provided with a graduated circle. In the mount- ing of the prism provision is made for adjustment in a vertical direction so as to ensure a wide field of view with all oculars. A slot in the collar in which the analyzer revolves engages a stud St on the draw-tube of the instrument. The draw-tube itself moves vertically only, thus if the polarizer and analyzer be properly inserted and their graduated circles set at zero, the prisms are crossed without further adjustment. The placing of the analyzer over the eyepiece in a microscope for microchem- ical analysis will be found to be much safer than the more con- venient mounting sliding into the body tube, as in petrographic instruments. When the instrument is to be much used in the microscopy of foods a supplementary polarizer may be obtained which fits into the ring below the Abbe condenser, thus allowing the prism to be swung quickly aside without interfering with the illuminating devices. Instruments made by other firms for chemical microscopy 22 ELEMENTARY CHEMICAL MICROSCOPY differ but little from that shown in the illustration. It has, therefore, been thought unnecessary to picture them here. Microscopes for Special Purposes. When large samples of powdered material are to be investigated, as in the examination of dry, powdered or granulated foods, drugs, etc., for adulter- ation, a microscope with large stage of the type shown in Fig. 4 FIG. 4. Microscope with Large Stage for the Rapid Examination of Powdered Material. is of great assistance. 1 The material is thinly spread out upon the plate glass stage, and the microscope is made to pass by means of the screws S and R over the entire area covered by the material. A very low power L is first employed until some particle is found, needing to be studied more carefully. The par- ticle is centered under the lens, L is then removed and the com- pound microscope M slipped in place in the same slot previously occupied by L. The particle in question now falls under the compound microscope. This type of microscope primarily in- tended for the examination of large sections of the brain will 1 Made by E. Leitz, Wetzlar and also by Nachet et Fils, Paris. MICROSCOPES FOR USE IN CHEMICAL LABORATORIES 23 be found a great saver of time, labor and material. Its appli- cations are many. In laboratory work involving the study of plates of bacterial cultures it will be found to be far superior to microscopes of the ordinary type, since plates of large size may be examined at any point within their areas. The compound microscope is provided with rack and pinion coarse adjustment and with a quick acting screw adapter F fitted to the end of the body tube for fine adjustment. Comparison Microscopes. It not infrequently happens that it is found desirable to carefully compare two preparations or two different samples. This is especially true in quantitative microscopy. With ordinary microscopes it is necessary to place first one sample, then the other, under the microscope, make drawings, measurements and take mental note of the appearance of each preparation in turn and then compare the mental pictures by the aid of the data at hand. This process is not easy, and the results not always trustworthy even in the hands of an expert without long and exceptionally thorough studies. Photomicrog- raphy offers a fair solution but here again the time required and the additional manipulations necessitated prevent its general application. This need of some device whereby quick and rapid compari- sons might be possible has long been felt, but no suitable instru- ments were placed upon the market until very recently. These new instruments have received the name Comparison Micro- scopes. They are so constructed that the images formed by two different optical systems are brought into juxtaposition, so that the observer is able to simultaneously see the images of two different objects. As long ago as 1885, Inostranzeff 1 employed what he desig- nated as a comparison chamber, consisting of two sets of totally reflecting prisms so mounted in a rectangular chamber as to reflect, into a single eyepiece, the images of half the field of each of two microscopes. Two years later Van Heurck 2 improved the Inostranzeff in- 1 Jahrb. f. Min., 2 (1885), 94; J. Roy. Micros. Soc., 1886, 507. 2 Van Heurck, J. Roy. Micros. Soc., 1887, 463. 24 ELEMENTARY CHEMICAL MICROSCOPY strument by a different arrangement of prisms. This latter type has again been revived by the Bausch and Lomb Optical Company in 1912, and by E. Leitz in 1914. A somewhat similar comparing device, consisting of two totally reflecting prisms, was proposed by Ewell 1 and employed by him as a colorimeter. The Van Heurck comparison eyepiece, Fig. 5, as constructed by Bausch and Lomb consists of a rec- tangular cell provided on the lower side with two orifices and FIG. 5. The Bausch and Lomb Comparison Eyepiece. with tubes T 1 and T 2 of the same diameter as ordinary oculars, and at such a distance apart as to permit their simultaneous insertion into the tubes of two microscopes placed side by side, Midway between these tubes on the top of the cell is an opening with a tube into which slides a Ramsden eyepiece O. Above the tubes T 1 , T 2 are placed totally reflecting prisms P 1 , P 2 , 1 Ewell, J. Roy. Micros. Soc., 1910, 14. MICROSCOPES FOR USE IN CHEMICAL LABORATORIES 25 which reflect the images, formed by the objectives of the micro- scope, into the rectangular prisms R 1 , R 2 , situated just below the ocular 0. The prisms R 1 , R 2 consist of rectangular pieces of glass cut through diagonally and cemented together, the inclination of the cut surfaces being parallel to the reflecting surfaces of P 1 , P 2 , respectively. Upon looking into the ocular O the field is seen to be divided into an upper and a lower part by a line passing from left to right. It is obvious that the image of half the field of one microscope will be seen in one of the halves of the ocular, while the other half of the ocular will exhibit half the field of the other microscope. In order to facilitate focusing the microscopes the tube T 1 is of such diameter as to fit snugly into the tube of one of the microscopes, while the tube T 2 is of less diameter and hence fits loosely. The microscope carrying T 1 is therefore focused first. Objects to be carefully compared by means of this instrument must necessarily lie in the same plane, otherwise the magnification in one half -field will be greater than in the other. Where slight variations in magnification can be neglected, the thicker preparation is placed upon the stage of the microscope carrying the tight tube of the comparison eyepiece, or if chemical microscopes (Fig. 3, page 20) are em- ployed, one or both preparations may be supported upon the auxiliary stage and turned down until the upper surfaces of the two preparations lie in the same plane. This, however, is only possible when no substage condenser need be employed. Comparison microscopes proper are of two different types, either they have a single eyepiece and make use of reflecting prisms or they consist of two microscopes with two eyepieces, the observer using both eyes. The Leitz 1 comparison microscope, Fig. 6, consists of two microscope tubes A, B attached to a single pillar P movable by rack and pinion. A single stage S is provided with two open- ings, one for each microscope tube. Under each stage opening is placed an Abbe condenser with iris diaphragm and rings for stops, or for blue, green or ground glass. Each condenser is illuminated by means of a separate mirror on a swinging bar and 1 Manufactured by E. Leitz, Wetzlar, Germany. 26 ELEMENTARY CHEMICAL MICROSCOPY is adjustable up and down by a friction collar. To the upper end of each microscope tube is attached a large chamber C, C 1 containing reflecting erecting prisms. Above the cham- bers are the oculars E, E 1 , provided with sliding dia- phragms D 1 , D 2 . The prism chambers are so constructed as to rotate through a small arc in the directions of the arrows, thus bringing the eyepieces nearer together or farther apart for adjustment of the proper pupillary dis- tance of the observer. The upper half of each eyepiece can also be rotated so that when the diaphragms D 1 , D 2 are inserted to cut off half the field in each ocular, they may be turned until the diam- eters of each half field are parallel or coincident. After turning through the proper arc the thumb screws T 1 , T 2 J. are tightened to prevent the adjustment from changing. By proper manipulation of the sliding diaphragms, the observer looking into the instrument with an eye above each ocular sees half the field from one preparation and half from the other in close juxtaposi- tion. A very rapid yet critical comparison of one preparation with another is thus easily accomplished. Or D 1 , D 2 may be so placed as to cut out the field of either tube, or if both are pushed in as far as they will go the fields will be superimposed, and the symmetry of two objects may be compared. FIG. 6. The Leitz Comparison Microscope. MICROSCOPES FOR USE IN CHEMICAL LABORATORIES 27 The coarse adjustment R by rack and pinion serves to roughly focus both tubes at once; then each objective is focused sepa- rately by means of the fine adjustment screw collars F, F 1 just above the objectives. That really satisfactory results may be obtained it is essential that both the sets of eyepieces and objec- tives shall be paired, i.e., shall have been constructed for use with a comparison microscope and be exactly equivalent in all properties. The fields are flat, brilliant, and with careful illu- mination and adjustment and a little practice most excellent results can be obtained. The instrument is adapted to all problems involving an exact comparison of size, structure or symmetry of microscopic objects, especially where the structure is so intricate as to render comparison and interpretation with the ordinary single compound microscope exceptionally difficult without recourse to photography. The value of the instrument in all problems of forensic chemical microscopy is evident. A second type of comparison microscope 1 is provided with a single eyepiece only, the field being divided into halves. As in the previously described instrument, two microscope tubes are attached to a single pillar and both focused together by rack pinion. Attached to the tubes is a rectangular closed chamber of the Inostranzeff type provided with two sets of totally reflect- in prisms, thus yielding to a single eyepiece half the field of view of each microscope. By means of a knob in the side of the chamber one set of prisms may be shifted at will so as to cut off the field of one instrument. In addition to a single fine adjustment, simultaneously affect- ing both microscopes, each tube is provided with independent fine adjustment collars just above the objectives. A single stage with two openings carries two substages, each with an Abbe condenser and with a mirror. The instrument may be employed with polarized light, thus affording exceptional opportunities for exact comparisons in the search for food adulterants and in microchemical analysis. Since in this in- strument we have a single ocular yielding a divided field, it is possible to obtain photomicrographs, half the area of the circle 1 W. and H. Seihert, Wetzlar, Germany. Thorner, Chem. Ztg., 36, 781. 28 ELEMENTARY CHEMICAL MICROSCOPY in the negative obtained being the image of one preparation, the other half that of the second preparation. This instrument consists essentially of a stand similar to the Leitz with the microscope tubes joined by a prism chamber and therefore no illustration of its construction is necessary. Photomicrographs and polarization studies are of course also possible with the comparison eyepiece described above. When two microscopes are available the comparison eyepiece will be found to perform all the work which may be accomplished by means of instruments of the Seibert type and will entail little additional expense to the equipment of the microchemical laboratory. Comparison microscopes are almost indispensable when fre- quent comparisons must be made between unknown and known or standard preparations, or when rapid approximate quantita- tive results are required. These instruments and the simple polarizing compound microscope may be said to be the only ones for what can be called general use in the chemical laboratory. Special microscopes for micrometric purposes, such as read- ing scales, determinations of the positions of lines in the photo- graphs of spectra, or measuring the diameter of depressions produced in testing for hardness by the Brinell method, will be found described in Chapter VII, page 147; microscopes for the study of ultramicroscopic particles in Chapter IV, page 54, while the special types of instrument for the examination of metallurgical products and large castings are taken up in detail in Chapter V. For the investigation of molten material, liquid crystals, etc., microscopes of special construction have in recent years been placed upon the market. Most of these have followed the de- signs of O. Lehmann and comprise a great variety of forms. 1 One of the simplest of these is shown in Fig. 7. In this instru- ment polarized light (see Chapter VIII) is obtained by reflection instead of by the usual manner by means of a Nicol prism, in order to permit swinging the tiny Bunsen burner B below the stage. The light rays reflected from P and R are polarized and 1 See Lehmann, Das Kristallisationsmikroskop, Braunschweig, 1910. MICROSCOPES FOR USE IN CHEMICAL LABORATORIES 29 are sent through the preparation upon the stage by means of the mirror M. The analyzer consists of a prism sliding in and out of the microscope tube at A. In the illustration the dotted FIG. 7. Simple Form of Hot-stage Microscope. Polarized Light is obtained by Reflection from the Plates P and R and the Mirror M as indicated by the Dotted Arrows. A = Analyzer. B = Small Gas Burner which swings under the Stage Opening. lines indicate the approximate direction of the light rays used to illuminate the object. When moderate temperatures are necessary the objective must be cooled by means of a blast of air directed upon the lower lens, and when high temperatures are employed the objective must be water-jacketed. CHAPTER III. ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES. Illumination and Illuminating Devices. Of even greater importance than the selection of the correct combination of objective and ocular for the study of a preparation is the matter of proper illumination. The earlier in his work the student appreciates the importance of illumination and the more thought and care he expends upon this phase of microscopic methods, the fewer errors he will make and the more easily will he accom- plish the objects of his investigations. For convenience of discussion the modes of illuminating objects for microscopic study may be grouped under the follow- ing heads: a. Transmitted axial light. b. Transmitted oblique light. c. Reflected axial light. d. Reflected oblique light. e. Oblique dark ground illumination. /. ''Orthogonal illumination" (Siedentopf Slit Ultrami- croscope). g. Differential color illumination. h. By means of ultraviolet light, thus causing certain sub- stances to become fluorescent. a. Transmitted Axial Light obtained by means of the mirrors with or without a condenser may be said to be the usual or most frequently employed method of illuminating transparent and translucent objects. With low power objectives and objects of coarse structure no condenser is necessary, but when the object to be studied presents a fine structure and delicacy of tracery and when its refractive index lies close to that of the mounting medium, structural studies become difficult, if not impossible, 30 ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 31 without moderately high powers and some form of substage condenser. It is therefore a safe rule to always employ a sub- stage condenser unless exceptionally low powers are to be used; this of course does not apply to problems involving examinations with polarized light. b. Transmitted Oblique Light is essential for the proper interpretation of appearances under the microscope of objects whose upper and lower surfaces are so placed as to lead to serious confusion if axial light is alone employed. Oblique light also aids in establishing whether the liquid medium or the object immersed in it has the higher refractive index. The value of oblique illumination may be better understood by referring to the diagram shown in Fig. 8. A transparent object O whose upper and lower surfaces are identical and perfectly symmetrical, is shown in section, lying upon an object slide upon the stage, with perfectly axial light as shown by the arrows. It will be obvious that even very careful focus- ing will fail to disclose the probable structure of the lower surface and that even the upper surface may be in doubt; but if oblique illumination be employed, usually a very faint shadowy image of the lower surface will be observed, slightly out of sym- metry with the upper surface. Swinging the mirror to one side or decentering the iris diaphragm of the condenser when this is possible, and noting at the same time any change produced in the image, will show that the image of the upper surface has the appearance of sliding over the lower, providing the objective has sufficient penetrating power. Under these conditions the trained observer is able to form a fairly accurate conjecture as to the morphology of the object under observation. Cleavage planes, infinitely narrow fissures or structures, the arrangement of whose elements is so fine and delicate as to be practically indistinguishable by axial light, may become easily discernible by oblique illumination; but as intimated above, ELEMENTARY CHEMICAL MICROSCOPY the character of the information thus gained is necessarily closely associated with the resolving power, penetration and, to a certain extent, the size of field of the optical combination above the stage. DEVICES FOR ILLUMINATION BY TRANSMITTED LIGHT. Condensers. In order that sufficient light may enter a high power objective to produce an image of such a degree of bright- ness as to be easily studied, it is essential that some device or apparatus shall collect, concen- trate and send through the object light rays at an angle which will fill the aperture of the objective. The usual construction of this device is shown in diagram in Fig. 9 and is known as the Abbe condenser. Condensers of this construction with two lenses have usually a numerical aperture, when employed to their full extent, of 1.20 and may be used with all ordinary dry objectives and with oil im- mersion objectives. They are designed to be used with the plane mirror. In the case of objectives of more than 1.20 N.A., a three or more lens com- Fxc. 9 . Diagra^of Abbe Condenser; ^.^ ^^ ^ ^ N.A. should be chosen. Con- densers used to their full aperture usually so flood the field with light, in the case of dry objectives, as to necessitate lower- ing them or closing their iris diaphragms or both until only just sufficient light rays are intercepted by the objective to fill its back lens and thus render the fine details of the illuminated object most distinct. ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 33 In the diagram, Fig. 9, the passage of the light rays is roughly indicated for a position of the Abbe condenser when used with an objective of low numerical aperture. The iris diaphragm is shown well closed. Usually it is advisable to also lower the condenser. Failure to employ the Abbe condenser in the proper manner or to appreciate the fact that a different adjustment is required to meet different prob- lems, is doubtless responsible for more errors in interpretation in microscopic examinations than any cause other than exces- sive magnification. Since very few dry achromatic objectives have a high numerical aperture it is evident that in order to obtain the best results it will be essential with all such optical combinations to close the iris diaphragm of the Abbe con- denser until the numerical aper- ture is no greater than that of the objective. It will be found to be a safe general rule to lower the Abbe condenser and to close FIG. 10. its iris diaphragm to a diameter about two-thirds or one-half that of the rear lens opening of the objective. The size of the diaphragm opening may easily be adjusted by removing the ocular, looking into the tube of the microscope and closing the diaphragm until the bright disk of light is reduced one-half or two-thirds. Oblique illumination with the Abbe condenser is quickest and most easily obtained by the method suggested by Wright of holding a finger below and half across the opening of the con- denser; the light rays then take the path roughly indicated in Fig. 10. Or we may drop upon the swing-out ring attached to the bottom of the condenser mounting a half-disk of black paper or cardboard, or a disk provided with a circular opening Diagram of Abbe Condenser; Oblique Light. 34 ELEMENTARY CHEMICAL MICROSCOPY to one side of the center. The disks furnished with the conden- ser, consisting of a central stop with narrow slots, yield very oblique illumination but a black background, and serve an en- tirely different purpose which is discussed elsewhere under the head Dark-ground Illumination. In the highest grades of mi- croscopes the substage mounting is arranged so as to provide a lateral movement of the iris diaphragm by means of rack and pinion. Oblique illumination is then obtained by closing the diaphragm to a small opening and racking it to one side. Oblique illumination is often essential to a proper interpre- tation of structure and to a sharp differentiation of refractive indices. The ordinary Abbe condenser is corrected for neither chromatic nor for spherical aberration and although it answers all the purposes of illumination in ordinary microscopy with standard objectives, in photomicrography or in combination with objec- tives of the highest grade and in work of the finest kind, its use is injudicious. Recourse should be had in such cases to achromatic or specially constructed condensers. Since investigations of this kind are rare in chemical laboratories, space forbids their consideration. In accurate crystallographic studies the microscope condenser must be especially free from both chromatic and spherical aber- ration; and instruments for this class of work are never provided with condensers of the Abbe type, but are always fitted with light-concentrating devices of special construction. It is essential that the optic axis of the condenser shall coincide with the optic axis of the microscope, or, in other words, the con- denser must be accurately centered. In the low-priced micro- scopes no provision is made for any adjustment of the mounting, the proper position being fixed by the manufacturer. Not infrequently through carelessness of workmen and inadequate inspection of the finished instrument, microscopes are sold whose substage condensers are so badly out of center as to render them unfit for high grade work. To test the adjustment of an Abbe condenser in a fixed mount- ing, close its iris diaphragm to the smallest obtainable opening, ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 35 raise the substage as far as it will go; insert a cross-hair eyepiece in the body tube and focus with a very low power upon the diaphragm opening. The diaphragm opening should fall at the center of the field of view directly under the cross-hairs, con- centric with their point of intersection. If the image of the opening is not centrally located there is something faulty in the construction of the condenser or in its attachment to the sub- stage, or in the alignment of objective and ocular. If the condenser has been found centered, we may change to a high power objective and be reasonably sure that the condenser will be centered with respect to the objective, providing a re- volving nose-piece is not in use; but if the objective is attached to an ordinary nose-piece, turning from one objective to another usually necessitates a readjustment of the condenser. With high powers, centering, as described above, is impossible and it will be found simpler to remove the ocular and hold a tripod or pocket magnifier over the tube; the image of the diaphragm opening is then easily seen and its relative position ascertained. In testing for proper centering it is important that the mirror be so placed as to yield exactly axial light. This may be assured by swinging the condenser to one side and placing upon the stage a preparation consisting of thin gum beaten up until full of air bubbles; a very tiny air bubble is selected and brought to the center of the field, it appears as a bright spot surrounded by a black ring; the bubble is sharply focused and the mirror adjusted by proper tipping until the bright spot appears exactly at the center of the circular black ring. The light is now exactly axial. This method of assuring absolutely axial light l is the simplest and surest available. Without touching the preparation or the mirror, carefully swing the condenser back in place, raise it about halfway and slowly raise and lower the body tube by means of the coarse adjustment, closely observing at the same time the appearance of the bubble image. If the light still remains axial with the condenser in place there will be no appreciable swaying of the image and no change of position of the bright spot of light. If 1 Gage, The Microscope, p. 48, loth Ed., Ithaca, 1908. 36 ELEMENTARY CHEMICAL MICROSCOPY the image sways and the bright spot of light is displaced to one side of the center the Abbe condenser is faulty and the character and the amount of the fault will be indicated by the magnitude of image displacement. In the better grades of Abbe condensers the mounting is fitted with centering screws, which permit moving the combina- tion of lenses so that the optic axis of the condenser lens becomes coincident with the optic axis of objective and ocular. The simplest method for easily centering adjustable Abbe condensers is to have a cap made, fitting exactly over the top lens of the condenser; at the exact center of this cap an exceed- ingly tiny hole is drilled falling in the optic axis of the apparatus. The microscope is focused upon this hole, illuminated by the light transmitted by the condenser and the bright spot seen is brought by means of the centering screws so that its center is coincident with the center of the field. It is the rule to always use the plane mirror with the Abbe condenser; but when the windows of a laboratory have small panes or wide cross bars it is often impossible to properly illu- minate an object with the plane mirror and Abbe condenser without projecting an image of the window bars into the field. Either the microscope must be moved very close to the window or the concave mirror must be used; the latter plan necessi- tates closing the iris diaphragm two-thirds or more and lowering the condenser. In aggravated cases a disk of ground glass may be placed below the condenser or in front of the mirror. The use of a disk of thin, fine ground glass will in fact be found a distinct gain in ordinary practice in the illumination of most objects. By its use softer, clearer and more easily interpreted images will often be obtained and the true colors of objects will be more easily recognized. When the recognition of colors becomes important, as, for example, in microchemical analysis, the student must remember that the image obtained in the microscope by illumination, by mirror and Abbe condenser with light from the sky will almost never show the true colors present in the object. To obtain a properly colored image, slide a piece of pure white ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 37 card at an angle of about 45 degrees between condenser and mirror, or place a disk of ground glass in the ring attached to the lower part of the condenser, thus obtaining, in part, reflected light and a gray or white background. The ring attached to the lower part of the condenser and arranged to swing aside serves to carry disks of blue glass to be employed when working with artificial light. By this means a much less fatiguing illumination is obtained, and providing the proper intensity of cobalt glass is at hand, white light giving proper color values is secured. Blue glass should always be placed below the condenser when working with yellow artificial lights. Most manufacturers supply blue glass disks with all their Abbe condensers. When the apparatus is to be employed in photography, yellow-green glass disks are furnished to be used as ray filters. c d. Reflected Light, Axial or Oblique, must be employed for the study of the surfaces of opaque objects or for the purpose of ascertaining the surface configuration of objects of any nature. In investigations of this sort the preparation may be illu- minated either by rays of light whose paths are oblique to the surface of the object and also to the optic axis of the microscope or by rays whose paths are parallel (or approximately so) to the optic axis and normal to the surface of the preparation. Oblique light rays are obtained either by means of small reflectors attached to the objective or by directing upon the object the rays from a radiant lying above the plane of the surface of the object. When a radiant is employed, as, for example, an arc lamp or a Nernst lamp, a condensing lens is usually inter- posed between light and object in order to concentrate the light rays and facilitate the proper placing of the illuminating beam. Illumination by a reflecting mirror may be obtained either by means of the mirror of the microscope, provided its swinging arm is long enough to allow raising the mirror above the plane of the stage, or by attaching to the objective a silvered metal paraboloid. The paraboloid illuminator was very popular at one time but has been almost entirely superseded by devices known as vertical illuminators (see page 76) in which the re- 38 ELEMENTARY CHEMICAL MICROSCOPY fleeting surface is mounted in a cell attached to the microscope just above the objective. In these devices the reflector sends the illuminating beam of light through the objective which acts as the condenser, concentrating the light rays into a bright spot of light upon the surface of the object at a point lying approxi- mately in the optic axis of the microscope. From the surface of the object the rays are reflected back through the objective and form the image of the object in the usual manner. When only very low powers are required for the examination of a specimen, simply holding it slightly inclined upon the stage will send sufficient light into the instrument to permit a thor- oughly satisfactory study of the coarse details. Slight focusing up and down will answer all purposes. Since reflected axial and oblique light must very frequently be employed by the chemist it is essential that he should thor- oughly understand the phenomena exhibited by different sur- faces illuminated in different ways. If we are dealing with a highly polished mirror surface S, Fig. ii (as, for example, a polished but unetched metallurgical specimen), lying in a plane normal to the optic axis of the microscope, and we il- luminate it by reflected light, it is obvious that none of the oblique rays ab, cd and ef can enter the objective to form an image since the angle of reflection is equal to the FIG. ii. Path of Oblique angle of incidence. The surface will there- Light Rays striking a fore appear fo,^ Tne more nearly a per- Plane Polished Surface. , J F feet reflecting surface the object possesses, the darker it will appear. It will remain dark until the ray ef becomes almost parallel to the optic axis and therefore prac- tically normal to the surface of S. Reflected light rays now can enter the objective and the surface appears bright and shining. But if the surface of the object illuminated by the oblique rays is irregular or etched, as diagramed in Fig. 12, then the irregularities will appear bright, the plane or polished surfaces dark. If a light ray a strikes a series of tiny minute points as at D, the light will be diffracted; diffraction patterns will be ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 39 formed in the field of the microscope and the true structure of the object at this point will prove very difficult of interpretation. When, however, axial reflected light is used, that is, when the illuminating beam strikes the polished preparation normal to its surface, the plane surfaces will appear bright, the irregularities FIG. 12. Path of Oblique Light Rays striking an Irregular Surface. FIG. 13. Path of Axial Light Rays striking an Irregular Surface. more or less dark, and minute projecting irregular points will yield diffraction patterns; for as shown in Fig. 13, the light rays b and c, striking reflecting surfaces, are turned aside at such an angle as to preclude their entering the objective. Careful consideration of the above described phenomena is absolutely essential to a correct interpretation of the structure of the material being studied. To determine when one is dealing with depressions and when with elevations when working with moderately high powers and vertical or oblique illumination is often a difficult problem which is further complicated for the beginner by the fact that the image seen is that of the object in a completely reversed position. It is obvious that the oblique illumination of opaque objects is restricted to low powers, since the free working distance of high power objectives is so small that the path of any pencil of light which will strike the preparation at a point lying in the line of the optic axis of the microscope must then be so oblique as to be approximately parallel to the surface of the preparation. Light rays reflected from the surfaces of anisotropic crystals are polarized, but are not noticeably polarized if from iso tropic crystals. It therefore often proves of great value in qualitative analysis to employ polarized light for the illumination of objects to be studied by means of vertical illuminators. 40 ELEMENTARY CHEMICAL MICROSCOPY e. Dark-ground Illumination is usually obtained by sending oblique light rays into the preparation from below, at such an angle that no rays directly enter the objective. This is accom- plished by introducing a metal stop below the Abbe condenser so as to shut out all central rays and allow only rays near the cir- cumference of the condensing lenses to enter the preparation, or, better, by substituting for the Abbe condenser a device which will reflect rays from a curved surface in such a manner as to bring them ^approximately to a focus. In preparations thus illuminated objects appear bright upon a black background. This method is invaluable for demonstrating the presence of very minute bodies or those whose index of refraction is so very nearly the same as that of the medium in which they occur as to cause them to escape detection when illuminated by trans- mitted light. It is generally the case that particles of a diameter of one micron or less require dark-ground illumination for their demon- stration. If the obliquity of the rays from the illuminating device is very great, the dark-ground illuminator becomes an " ultra- condenser " and may be employed for demonstrating the presence of particles less than 0.2 /i in size. Dark-ground illumination is employed in practice in the ex- amination of blood for the presence of parasitic organisms, in the study of bacteria, in the biological examination of water, in the study of foods, fibers, crystallization phenomena, tiny crystals, submicroscopic particles, colloids, etc. If the Abbe condenser is to be employed for dark-ground illumination, insert one of the dark ground stops in the ring attached to the bottom of the condenser mounting, open the iris diaphragm to its full capacity, and screw up the condenser in its mounting until, when turned in place and the substage is racked up to its highest point, the upper lens will just touch a slide laid upon the stage. A drop of water is then placed be- tween the condenser lens and the preparation to be examined. It is always essential to ascertain the thickness of object slides which yield the best results and keep this value for future refer- ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 41 ence. Special dark-ground illuminators are marked by the manufacturers with the thickness of object slide for which they are designed. The use of an Abbe condenser with dark-ground stop as a substitute for special dark-ground illuminators is not to be recommended since the obliquity of the rays is seldom sufficient to prevent some light from entering the objective. The results usually obtained are likely to be poor and unsatisfactory. Dark-ground Illuminators are condensers of such construction that very oblique light rays are caused to converge, usually by reflection, and to so strike the lower surface of a cover glass placed over the preparation to be studied as to be totally re- flected. To prevent axial light from passing through the illu- minator an opaque stop is placed in the optic axis of the device. Cover Glass FIG. 14. Path of Rays in Reflecting Condenser for Dark-ground Illumination. The field is therefore black or nearly so, save for a slight halo at its edges, while the objects appear bright or brilliantly colored upon a dark background. In Fig. 14 a simple paraboloid reflecting illuminator is shown diagrammatically in section, with the directions of the light rays so exaggerated as to make clearer the reason the field of view is dark. ELEMENTARY CHEMICAL MICROSCOPY Sections of typical illuminators are shown in Fig. 15, A, B, C, D. It will be seen that although the construction may be different in different types, the rays emerge at approximately similar angles. In illuminators of these types (B, C, D) the curvatures of the reflecting surfaces are ground after mathe- matically calculated curves which will bring the light rays ap- proximately to a focus at a point upon the cover glass. In the FIG. 15. Types of Dark-ground Illuminators. A. Nachet et Fils. B. Reichert. C. Bausch & Lomb. D. E. Leitz. diagrams for simplicity, cover glasses and preparations have been omitted. The cheaper forms of dark-ground illuminators fail to bring the rays to a true focus and instead of a point of light upon the cover glass we obtain a disk, as shown in an exag- gerated manner in Fig. 14. An exception to the above statement, relative to the construc- tion of reflecting condensers, is found in the Beck 1 dark-ground 1 Made by R. & J. Beck, London. ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 43 illuminator in which, Fig. 16, a lens is combined with a paraboloid to bring the rays to a proper focus. The Beck illuminator is unique in that it permits adjustment for different thicknesses of object slides, an impossibility with other forms of paraboloid illuminators. ^ This adjustment for slide thickness is accomplished by changing the distance between the focusing lens L and the pa- raboloid P. As seen in the diagram, the illuminator consists of two parts, the paraboloid mounting screwing into that which holds the lens; therefore raising or lowering the paraboloid will displace the focal point / and bring about an accom- FlG - l6 - Beck Adjustable modation for different thicknesses of slides. Dark -g round illuminator. In practice it is rarely possible to have such accurate grinding that all the rays are properly deflected and none enter the objec- tive. Only those rays included in a low numerical aperture are available. Hence the employment of an objective of high nu- merical aperture and very short working distance yields a field which is never dark. Since practically all high power immersion objectives are made with as high numerical apertures as possible, it is absolutely essential that some means be used to reduce their numerical aperture below i, if they are to be employed in dark-ground studies. This is accomplished by introducing into the objective mount some form of diaphragm : or specially con- structed objectives of N.A. less than i may be purchased. Diaphragms for use with objectives in dark-ground studies are generally supplied by the manufacturers of reflecting condensers for introduction into the special objectives to be used. These funnel-like diaphragms are not interchangeable and can be employed only for Jthe objective for which they are designed. Figs. 17, 1 8 and 19 show three different types and forms of dia- phragms employed for this purpose. In the case of Fig. 17 the lens mounting is unscrewed just back of the back lens combina- tion and the funnel diaphragm, provided with male thread, is screwed into the opening tapped into the upper half of the objec- 44 ELEMENTARY CHEMICAL MICROSCOPY tive mounting. In the case of Fig. 18, the objective is also unscrewed just above the back lens combination, but in this case the diaphragm is merely dropped into the hole in the lower half of the mounting, while in the case shown in Fig. 19, the long tubular diaphragm is inserted into the objective from above FIG. 17. FIG. 18. FIG. 19. Methods of Reducing Numerical Aperture of Objectives for Dark-ground Studies. (D, D, D, Removable Diaphragms.) without necessitating any separation in the mounting of the objective lenses. By means of these diaphragms the numerical apertures of the objectives are reduced to approximately 0.9 or 0.95. In order to obtain the maximum resolving power with dark- ground illumination Conrady has shown 1 that the condenser must have not less than three times the numerical aperture of the objective. He suggests that the practical resolving power obtainable may be expressed as equal to J N.A. objective + i N.A. condenser, but Reinberger points out that on actual trial 2 the Conrady formula gives results about 25 per cent too low. The inexperienced observer, however, will find that the resolving power obtainable in his work will conform rather closely with the Conrady formula. It is therefore well to bear in mind that in dark-ground illumination studies fine details of structure are to be discerned only with the greatest difficulty 1 Conrady, J. Quekett Micro. Club, 11 (1912), 475. 2 Reinberger, J. Quekett Micro. Club, 11 (1912), 503. ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 45 and will require extreme care in adjusting the illumination and in selecting the proper objectives. 1 It is evident that with a properly selected optical combination, the field of view will appear black or very dark, while any objects present will appear to be bright and self-luminous. The more oblique the rays the more minute the particles may be whose presence will be revealed by their diffraction patterns. When the upper limit of obliquity is reached the illuminators are usually designated as ultracondensers and the instruments to which they are attached are then known as ultramicroscopes. There is no sharp dividing line between ordinary dark-ground illumination and ultramicroscopic illu- mination; the one gradually merges into the other. In all ultra- microscopes we are dealing with dark-ground illumination, but, on the other hand, few dark-ground illuminators yield light rays sufficiently oblique to demonstrate particles of ultramicroscopic FIG. 20. Types of Reflecting Condensers for the Study of Ultramicroscopic Particles. size. Typical ultracondensers are shown in Fig. 20. A com- parison of the indicated light ray directions in these with those in Fig. 15 will disclose that their inclination is considerably greater. For the chemist the ultracondensers are of far more value than simple dark-ground illuminators and those fitting into the substage will be found preferable to those of plate form, 1 Siedentopf and Zsigmondy have shown (Ann. d. Phys. [4] 10 (1903), 14) that in the ultramicroscope the brilliancy of the diffraction disks is proportional to the product of the squares of the numerical apertures of the image-forming and illu- minating objectives. 4 6 ELEMENTARY CHEMICAL MICROSCOPY e.g., Fig. 21, which lie upon the stage of the microscope. More- over, all ultracondensers can be employed as ordinary dark- ground illuminators, the only drawback in routine work being that they require more careful adjustment. s \ FIG. 21. Simple Dark-ground Illuminator for Use upon the Stage of the Microscope. The Adjustment of Dark-ground Illuminators for use requires close attention, chiefly, to four conditions: (i) a selection of a sufficiently powerful radiant and the projection of a spot of light large enough to completely fill the lower opening of the illuminator; (2) the employment of objectives having a numer- ical aperture never greater than i.o; (3) the use of object slides of the thickness for which the illuminator has been designed; (4) accurate centering of the illuminator with respect to the optic axis of the microscope. An examination of the diagrams (Figs. 15 and 20) will show that theoretically the oblique rays meet to form a tiny spot of light just outside the apparatus in the line of its optic axis. It is obvious that this spot should lie in the optic axis of the objec- tive and the ocular. In order to facilitate centering, a tiny circle is usually engraved upon the upper surface of the glass of the illuminator; this circle is focused with a low power and is brought to the center of the field of the microscope, either by means of centering screws c, c, Fig. 22, provided for this pur- pose, or is moved by the fingers when a stage illuminator, Fig. 21, is placed upon the stage. If the microscope is provided with ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 47 a revolving nose-piece the objective used in centering should be removed and the high power to be employed in the dark-ground studies substituted in the same opening in order that there shall be no change in the relations of the optic axes. When employ- ing ultracondensers of the highest type it is better to remove the nose-piece and to attach to the body tube a centering adapter into which the objective is screwed; this permits accurate cen- tering of each objective used and therefore much better optical conditions are obtainable. d FIG. 22. Paraboloid Dark-ground Illuminator for Use below the Stage. In order, however, that the objective may be centered, it is essential that we have a central fixed point upon the stage to which we may refer. Stands to be employed for high-grade ultramicroscopic work should be provided with mechanical stages with graduated coordinate motion and a centering object slide, carrying at its center a tiny cross. When placed upon the stage so that the different scales of the mechanical stage occupy the positions which the manufacturer has indicated upon the object slide, the point of intersection of the ruled cross will fall exactly in the axis of the tube of the microscope. The objective is focused sharply upon the cross and if the center of the cross does not fall in the center of the field it is brought there by moving the screws a, a, Fig. 24, page 59. If the condenser is not provided with an engraved circle upon its upper surface it may be centered by placing an object slide upon the stage with immersion fluid, usually water, between it and the condenser; the light spot from the radiant is next prop- erly adjusted and the mirror inclined until a bright spot of light appears upon the object slide. The condenser is raised or lowered 48 ELEMENTARY CHEMICAL MICROSCOPY until the spot of light attains its smallest size. Focus upon this tiny spot with a low power objective; if the condenser is properly centered the spot will lie at the center of the field. Should it lie to one side, bring it to the center by means of the centering screws or center the objective with respect to the point of light. Having adjusted the condenser, the next step, if the device is of the cardioid type (see page 67), is to ascertain whether the quartz cell, which must be used with the instrument, is in proper condition for use. Lay the quartz cover upon the cell and press it down very carefully. Notice whether there appears at the zone of contact between cell and cover a series of colored concentric rings. If the pattern does not consist of concentric circles, but appears to be elliptical, it is probable that the cell is not level with respect to the optic axis. Adjust the level screws until the plane of the cell is normal to the optic axis. If the eccentricity of the rings does not disappear, the trouble lies in the objective which is not corrected for the thickness of the cover of the cell being used. A powerful source of light is essential. Direct sunlight by means of a clockwork heliostat is ideal but seldom available. The next choice is an electric arc of 4 to 5 amperes or more, for ordinary dark-ground examinations, and of 15 to 20 amperes for ultramicroscopic studies of colloids, etc. Useful types of radi- ants will be found described on page 132. The more powerful the radiant the smaller the particles which can be demonstrated. Siedentopf estimates that direct sunlight will reveal the presence of particles whose diameters are one-thirtieth of that of the smallest appreciable with the ordinary arc lamp. Since the light rays enter these reflecting condensers through an annular space, there being an opaque stop at the center, it is obvious that the spot of light reflected from the mirror of the microscope must have a diameter slightly greater than this space, otherwise the illuminator will not properly function; for this reason, before placing the illuminator in position for cen- tering, it is always essential to examine its lower surface and ascertain the diameter of the spot of light necessary to completely ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 49 fill the annular entrance space. The radiant and a suitable condensing lens are then so placed as to yield parallel rays and produce a spot of light of the proper size and intensity at the center of the plane mirror of the microscope, the mirror being so inclined as to reflect the light rays into the dark-ground illu- minator. Dark-ground illuminators require that an immersion fluid be placed between them and the object slide. Usually freshly filtered water is sufficient, although homogeneous im- mersion oil sometimes yields better results, especially with illu- minators of the plate type. In applying the immersion fluid and laying the object slide in place great care must be taken to prevent the entrance of air bubbles or dust particles. Because the light rays are caused to emerge from the illumina- tor at such an angle (determined by the inclination of the reflect- ing surfaces) as to converge to an axial point lying just above the plane of the object upon the object slide, it is, of course, essential that the thickness of the object slide be known, for if too thin the illuminating rays will meet too far above the material to be studied, or if too thick the focal point will lie too low; for these reasons optical instrument makers mark upon the de- vices the object slide thickness to be employed. For example: Thickness of object slide. Bausch and Lomb paraboloid illuminator Zeiss paraboloid condenser Reichert reflecting condenser Reichert slip-in reflecting condenser Leitz reflecting condenser Zeiss cardioid condenser for quartz cell i .40 to i .55 mm. i .o to i . 10 mm. 0.7 to 1. 10 mm. 2.0 mm. less than i mm. i .2 mm. Absolutely clean object slides and cover glasses are essential and great care must be exercised in wiping off the immersion fluid from the condenser to avoid scratching the glass. Lens paper of the highest grade only should be employed, and the wiping off of the fluid should be done with the least pressure possible, otherwise fatty material from the fingers may be forced through the pores of the lens paper upon the glass. A mere 50 ELEMENTARY CHEMICAL MICROSCOPY trace of grease upon the glass surface will lead to the formation of air bubbles, or will prevent optical contact if water is the im- mersion fluid. The preparation to be studied must be thin and must be covered with exceptionally clean and very thin cover glasses. Covering the preparation with a cover glass is essential. In order to expedite the adjustment it is well to have at hand a permanent slide of some material which yields good results with dark-ground illumination, as, for example, diatomaceous earth. With such a preparation on the stage the radiant, mi- croscope mirror and the condenser are all so mutually arranged as to yield the best illumination of the diatoms; the final adjust- ment is then made by raising or lowering the condenser. The test slide may now be replaced by the preparation to be studied. Little change, if any, should be required to give the most sat- isfactory results. If material of unknown structure or com- position is placed upon the stage without a prior examination of material of known behavior much time may be lost in attempting to interpret anomalous appearances due to improper illumination. Owing to the exceedingly complicated diffraction patterns often obtained with dark-ground illumination great difficulty may be experienced in arriving at a correct explanation of the phenom- ena observed, and it is only after study of materials of known structure that it is safe to proceed to examinations of somewhat similar material of unknown structure. /. Orthogonal Illumination is a term applied by Zeiss after Siedentopf and Zsigmondy to an arrangement of radiant, con- densing lenses and tiny slit such that the light rays enter the preparation at right angles to the optic axis of the microscope. The presence of particles is thus indicated by the light diffracted from them, the particles themselves remaining invisible and only the diffraction patterns, which may be relatively large, are seen in the field of view. This mode of illumination, as well as that by exceptionally oblique rays, given above under e, applied to microscopic examinations gives us instruments commonly called ultramicroscopes . ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 51 Orthogonal illumination is employed in the study of colloids and other particles in suspension in liquids and for the study of particles in transparent or translucent solids, such as glass, etc. For details as to apparatus and their use, see page 57. g. Differential Color Illumination by the method of Rhein- berger 1 may be obtained by substituting for the dark-ground wheel stop of the Abbe condenser colored disks of transparent material, using a darker color for the central portion and sur- rounding this disk with an annular ring of a lighter and strongly contrasting color. The object will then appear strongly illu- minated, but colored upon a colored background. If, for exam- ple, the central disk is blue and the ring red, the objects will appear red upon a blue background. With care and a suitable choice of colors, very remarkable results may be obtained which may greatly facilitate the study of certain sorts of material. h. By Means of Ultraviolet Light. When ultraviolet rays impinge upon certain substances they become fluorescent and glow with violet, red, green or bluish light. The color of the fluorescence is peculiar to the substance. Since comparatively few bodies exhibit this phenomenon and since the color is a further aid in differentiation, advantage has been taken of this property of bodies as a means of identification of such sub- stances not readily recognized when present in low per cents hi mixtures. To permit the extension of this method to minute amounts of material the " Fluorescence Microscope" has been constructed. 2 Ordinary glass is practically opaque to ultraviolet rays but not to the light rays resulting from the fluorescing of the sub- stance; the ultraviolet rays however readily penetrate quartz. We have, therefore, only to substitute quartz for glass in the condenser in order to concentrate the ultra rays on the object upon the stage. It follows from this that although the illu- minating devices must be of quartz, as also the object slide upon which the object lies, the objective and ocular may be those ordinarily employed. 1 J. Roy. Micro. Soc., 1896, 373; Spitta, Microscopy, London, 1909; 175-178. 2 Made by C. Reichert, Vienna, Austria. 52 ELEMENTARY CHEMICAL MICROSCOPY Either a carbon arc with special carbons or a mercury vapor lamp may be employed as radiant. Fig. 23 shows diagrammatically the construction of a fluo- rescence microscope. The rays from the radiant R are concen- trated by the quartz condensing lens Q, then pass through the Wood-Lehmann filter F consisting of a quartz or of a blue "Uviol" glass cell, thence the rays pass to the reflecting quartz X FIG. 23. Reichert Fluorescence Microscope. prism P which in turn reflects them into the quartz lens dark- ground condenser. This device brings the ultraviolet rays to a focus upon the object supported upon the stage by means of an object slide of quartz or of Uviol glass. Ordinary glass, besides being practically opaque to rays of very short wave length, as stated above, fluoresces with a violet or bluish tint under the action of the ultraviolet rays and cannot therefore be employed as a support. If it is necessary to cover the preparation ordi- nary glass cover glasses may be employed, but it is unwise to do so if thin quartz cover glasses are available. As in all dark-ground illuminators, an immersion fluid between condenser and object slide is essential. In this case glycerine is employed (n = 1.47). The light filter whose function is the removal of waves of long wave length, affecting the eye as light, consists of two com- partments, one filled with a 20 per cent copper sulphate solution, the other with an aqueous solution of nitrosodimethyl aniline (i: 12000). The only changes in construction and materials lie entirely ILLUMINATION OF OBJECTS; ILLUMINATING DEVICES 53 in the illuminating devices. Any microscope permitting the attachment of a dark-ground illuminator whose lenses are made of quartz may be converted into a fluorescence instrument. Although this system of illumination is still so new as to have been tried by but very few workers, its future development seems assured and its usefulness in qualitative chemical analysis of minute fragments of material to be unquestioned. 1 It is valuable not only in the analysis of inorganic material, such as crushed minerals, soils, mixtures of tiny crystals, etc., but is of equal value in organic analysis, in the examination of foods for adulteration and even in the microscopy of drinking water. 1 See Heimstadt, Das Fluoreszenz-Mikroskop, Zeit. f. wiss. Mikros., 28 (1911), 330; Wasicky, Das Fluoreszenz-Mikroskop in der Pharmakognosie, Pharm. Post, (1913); Lehmann, H., Das Lumineszenz-Mikroskop, seine Grundlagen und seine Anwendungen, Zeit. f. wiss Mikros., 30 (1913) 417. CHAPTER IV. ULTRAMICROSCOPES. APPARATUS FOR THE STUDY OF ULTRAMICROSCOPIC PARTICLES. Ultramicroscopes. Attention has already been called to the fact that the compound microscope with transmitted axial light will resolve tiny particles in suspension in a liquid only when there is a certain appreciable difference between the re- Iractive index of the particles and that of the liquid, and when the diameters of the particles are greater than half the value as- signed to the shortest wave lengths producing the effect of light upon the normal human eye. We have also seen that if instead of axial light, oblique rays are employed the ability to discern minute particles and intricate structure is greatly increased, especially if the obliquity of the rays is such as to yield an illu- minated object upon a black background. If the degree of inclination of the illuminating rays be still further increased and the source of the rays a powerful radiant and the objective employed one of low numerical aperture, only light diffracted by the object will enter the objective; the phenomenon known as the a Tyndall effect" results, so familiar in the scintillating dust particles visible when a ray of sunshine enters a tiny opening in a darkened room or cell. The existence of these infinitely minute particles in suspension in the air is manifest to the naked eye through that phenomenon, although even a high-power microscope fails to resolve them. The ultramicroscope is merely the adaptation of this Tyndall effect to microscopic illumination. As a result, the existence may be demonstrated of particles almost one thousand times smaller than is possible by means of the most powerful instrument employed in the usual manner. It is obvious that under the illumination of these very oblique rays, light alone which has been diffracted or reflected by the 54 ULTRAMICROSCOPES 55 particles enters the microscope and eventually the eye of the observer, and that therefore he never sees the particles them- selves, but merely a diffraction disk of light. We know of the existence of these particles through the same manifestation of more or less scintillating points of light that we see in the fixed stars on a moonless night. As hereinbefore stated the image of a point of light is a diffraction disk surrounded by alternate dark and bright rings. These diffraction disks appear to be in rapid motion. They appear to spin, to expand or contract and are endowed with a constant vibratory movement. This is due to the fact that exceedingly minute particles suspended in a liquid exhibit a constant vibratory and rotatory motion, long called the Brownian movement and now known to be as- sociated with and a manifestation of what we commonly term molecular vibration or bombardment. The presence of disin- tegrating or so-called " digestive " colloids increases the Brownian motion, while electrolytes by reason of their causing agglutina- tion tend to decrease the amplitude of the paths of vibration. In the few years that ultramicroscopic research has become possible a large number of investigations have been made upon the amplitude of the paths of vibration of the finest of these infinitely small suspended particles, with the result that the measurements made agree very closely with the theoretical values computed for the amplitudes of vibration of the molecules. Agencies which increase molecular vibration, such as heat, dilution and consequent reduction of viscosity, increase the Brownian movement. Hence, we find under the ultramicroscope the suspended particles in a gas (as, for example, in smoke) in much more rapid motion than in a liquid, while in a solid the Brownian movement is visible only with the greatest difficulty. Since the tiny particles in suspension are being bombarded on all sides, the motion imparted to them must be the resultant of the forces acting ; we therefore find them spinning rapidly as well as moving to and fro. Some authors have even suggested that the term kryptokinetic motion be assigned to the rotatory movement to distinguish it from the oscillating Brownian vi- bration. 56 ELEMENTARY CHEMICAL MICROSCOPY The amplitude of the Brownian movement may be ascertained by means of a net ruled eyepiece micrometer calibrated in the usual manner. Space forbids a discussion of the experimental details. (Note. The student should read the excellent summary of the then known facts relative to the Brownian movement given by Rutherford in Science, 30 (1909), 289-302.) The light emanating from the particles is polarized, the in- tensity of polarization increasing with the decreasing size of the particles. This fact enables us to differentiate between light diffracted by the particles and light emanating from flu- orescent bodies, since fluorescent light is not polarized. A well- equipped ultramicroscope must therefore include a device for the projecting of polarized light into the preparations and an analyzer for the study of the light rays forming the image in the microscope. But it must be remembered that even in the highest developed types of the ultramicroscope tiny particles in suspen- sion are discernible only when the refractive indices of these particles are different from that of the medium in which they are suspended; otherwise, no light will be diffracted from them. Therefore, although a medium may appear to be " optically empty" when viewed in the ultramicroscope, it by no means follows that there are no so-called " colloids" in suspension. To meet this difficulty and to extend the range of the ultrami- croscope, W. Ostwald 1 has suggested that monochromatic light be employed. This suggestion is based upon the fact that although two substances may have an identical value for their refractive indices for white light, with light rays of certain definite wave length the indices may be sufficiently different to permit the illuminating rays to render the tiny particles manifest. To the smallest particles visible in the ultramicroscope the terms micellae, ultramicrons or submicrons are sometimes given. Particles still smaller and therefore invisible in the ultra-micro- scope are called amicrons. The earliest practical instrument may be said to be the Slit 1 Ostwald, W., Zeit. f. Ind. Kol., 11 (1912), 290. ULTRAMICROSCOPES 57 Ultramicroscope of Siedentopf and Zsigmondy. At first sight this instrument might be thought to be also the most efficient, in that the path of the illuminating rays entering the object cell is at right angles to the optic axis of the observing microscope; but it must be remembered that owing to internal reflections and the impossibility of obtaining a perfectly black background the field is never sufficiently black to render very feeble diffrac- tion evident. This failure to obtain a black background is due, as first stated, to internal reflection on the one hand and upon the other, to the fact that the beam of light entering the cell is usually of such a diameter that when the objective is focused upon it there is always a plane below that in focus which con- tains bright particles. Moreover, this trouble is aggravated for the reason that it is essential to use objectives of long working distance and great penetrating power. These difficulties are largely eliminated in the more recently perfected ultracondensers of the dark-ground illuminator types, since in these devices not only is the background blacker but the light entering the liquid under observation is greater in quantity. For example, in the cardioid condenser, 1 the makers estimate that its light-concen- trating power is approximately twenty times that of the slit ultramicroscope. In spite of this advantage of the ultracondenser to demon- strate the presence of particles in suspension greatly beyond the limit of instruments of the slit type, preference should be given to the latter form for general use in the chemical laboratory when only a single type of instrument can be purchased, because of the fact that the slit microscope is universal in its application, serving equally well for solids, liquids, gases or vapors, and for hot or cold preparations, while the reflecting condenser types are confined to the study of thin films of liquid at room tempera- ture (or in certain restricted cases to the study of tiny transparent fibers). 2 The Slit Ultramicroscope consists of an ordinary compound microscope, a special cell of black glass with small windows at 1 Made by Carl Zeiss, Jena. 8 Gaidukov, Zeit. angw. Chem., 21, 1 (1908), 393. 58 ELEMENTARY CHEMICAL MICROSCOPY right angles to one another and an illumination device for pro- jecting a tiny beam of light into the cell in a line at right angles to the optic axis of the microscope. The tiny beam of light is obtained by means of small projection lenses and an adjustable slit. To distinguish this type of illumination from others com- monly employed in microscopy, the term " orthogonal illumina- tion" has been proposed. It is obvious that in this system no direct light can enter the objective but only such rays as are diffracted by the particles in suspension in the liquid contained in the cell. The form and arrangement of the component parts of the slit ultramicroscope naturally differ according to the optical firm manufacturing the instrument. One of the best known and most frequently used types is that shown in Fig. 24. 1 This instrument consists of an optical bench B, at one end of which is placed an arc lamp R and at the other a compound microscope. Between the lamp and the microscope there are a series of con- densing lenses and an adjustable slit. The light rays emanating from the arc are collected by the spherically and chromatically corrected lens Ci of 80 millimeter focus, so placed as to project a very bright image of the crater of the arc upon the slit S. In ordinary use this slit has its length in a horizontal position, the width being controlled by the micrometer screw with graduated head G, while the length of the slit is regulated by the screw s. After passing through the slit the light rays enter the lens 2, having a focal length of 55 millimeters, whose function is to project a reduced image of the slit into the condenser-objective C 3 . Since both slit and lens 2 are movable forward and back upon the optical bench, the lens C% serves a double purpose, projection and adjustment of the magnitude of the light beam entering Ca. The objective Cs projects into the preparation contained in the cell of black glass U a tiny conical beam of light at right angles to the optic axis of the microscope M. To prevent any side light from entering the preparation, lenses Ci and 2 are small and are mounted in blackened metal screens; as a further precaution a large metal screen D with tubular 1 Manufactured by Carl Zeiss, Jena. ULTRAMICROSCOPES 59 tjxy 60 ELEMENTARY CHEMICAL MICROSCOPY opening or adjustable diaphragm is introduced between the radiant and d. The objective C 3 screws into a tube fitting into the sleeve T and may be slid forward and back for coarse ad- justment. A very sensitive forward and back movement is further provided by the fine adjustment screw Vi. A second fine adjustment to the right and left for accurately centering the illuminating cone of light is obtained by the screw V 2 . By means of these two screws it is possible to adjust the tiny beam entering the material to be studied, in such a manner as to ensure the focal point of the condensing objective C 3 falling in the line of the optic axis of the observing microscope M, and therefore have the whole of the tiny beam lying across the exact center of the field of view. FIG. 25. Biltz Cell. To the lower end of the body tube of the microscope is at- tached an adapter A with centering screws 0, a, providing a device for accurately centering the objective O (see page 47). The liquid containing suspensoids is conveniently placed for examination in a Biltz cell, Fig. 25, or, when the short piece of FIG. 26. Biltz-Thomae Cell. rubber tubing which is attached to the end of the tube is objec- tionable because of its possible action on the colloids, a Biltz- Thomae cell, Fig. 26, may be substituted. In both of these cells the essential feature is the central dark glass chamber of ULTRAMICROSCOPES 6l about 3 millimeters internal diameter, provided with two small windows at right angles to each other these two windows consist of either thin glass or, better, of very thin quartz disks cemented in place. The passage of the beam of light through one of these cells is shown in the diagram, Fig. 27. No light other than that diffracted from the particles in suspension in the liquid can enter the observing microscope. The cell is usually attached to the microscope objective by a special cell holder; FIG. 27. Illuminating Rays in the Cell of the Slit Ultramicroscope. this, however, is open to the serious defect of difficulty in focus- ing and that cells purchased at different times are not exactly of the same thickness of wall, and hence the center of the upper window will not fall in the optic axis of the microscope. For these reasons the author prefers to support the cells upon an elevating mechanical stage P, as shown in Fig. 24. This arrange- ment permits the shifting and easy adjustment of the cell, so that its upper window is exactly centered with respect to the optic axis of the observing microscope. The cell is held in place by the spring clips c. The stage supporting the cell U may be raised and lowered by means of a knurled nut q. The nut p clamps the stage in place while the screws Wi and W^ serve to move P forward or back and to the right or left. One of the most serious defects of the Biltz cell is the difficulty of properly cleaning it after use, especially when there has been deposition of a colloidal film upon the windows. Treatment with a proper solvent and long washing is imperative. Before introducing a liquid for examination it is always best to pour a little alcohol through the cell and to follow this with the alco- holic solution to be studied, or if aqueous suspensions are to be 62 ELEMENTARY CHEMICAL MICROSCOPY employed, displace the alcohol with distilled water free from all fatty or greasy matter and then introduce the colloidal solution. This process is usually essential in order that the liquid to be examined shall come into perfect contact with the windows of the cell with no interfering film and no air bubbles. A much cheaper and simpler cell is shown in Fig. 28. 1 It consists of a tube of black glass with central swelling and win- dows at right angles to each other. These win- dows are either of glass or of quartz, the latter being preferable, since glass is slightly fluores- cent. For use, two pieces of rubber tube are at- FIG. 28. Simple Cell for Use with Slit tached as shown by the Ultramicroscopes. dotted Hnes Thege j-^ cells give excellent results with gases and vapors and may also be employed for the study of such solutions as will not be affected by contact with rubber. For preliminary examinations they are far more convenient than the Biltz cell and like it can easily be held in place on the type of stage shown in Fig. 24 by thin metal clamps or rubber bands. Moreover, these cells are more easily cleaned and are relatively inexpensive. When solids are to be examined, as, for example, specimens of glass, it is important that there be two sides of the preparation which meet at as nearly right angles in as sharp an edge as is possible. The reason for this will readily be understood by re- fering to the diagram, Fig. 29. If the sides do not meet in a sharp edge as shown at a, but form an obtuse angle or rounded edge b, the beam of light must be lowered below b. If this is done, the beam of light R will lie too low to be focused, even if the lower lens of the objective is brought into actual contact with the upper surface of the object. In this case the beam lies beyond the working distance of the objective. Should we at- tempt to bring R within the range W, as indicated in the lowest 1 Made by E. Leitz, Wetzlar. ULTRAMICROSCOPES diagram, diffraction, refractions, reflections and dispersions take place of such characters and to such degrees as to render the detection of micellae impossible. No suggestions as to optical combinations or size and intensity of the illuminating light beam may be given which will be applic- able to all materials. As in all other cases of microscopic investi- gation, the proper conditions must be experimentally ascertained for each preparation examined, but it is a safe rule to always avoid too large a slit and too high a magni- fication. For the slit ultramicroscope as made by Zeiss two objectives are specially constructed, a dry 7 mil- limeter, 0.4 N.A. achromatic ob- jective for the study of solids, and a 4.4 millimeter water immersion of 0.75 N.A., for use with cells containing solutions. A good gen- eral outfit should include oculars, i, 6, 8, 12 and 18. When polarized light is necessary in the study of colloidal reactions a nicol prism as polarizer mounted upon a saddle stand is placed be- tween the lens Ci and the slit S. The analyzer is then placed as usual above the ocular of the mi- croscope M. To adjust the illuminating beam of light used with the slit ultramicroscope shown in the diagram, screw the condenser- objective Ca into its holder T. Place the projection lens 2 at about 10 to 12 centimeters from the end of T, place the adjustable slit approximately 12 centimeters from 2, the projec- tion lens Ci about 12 to 15 centimeters from the slit, the dia- FIG. 29. The Necessity of having Two Sides at Right Angles in the Object for Ultramicroscopic Study. 64 ELEMENTARY CHEMICAL MICROSCOPY phragm D 12 to 15 from Ci and the arc lamp so that its carbons are about 8 centimeters from D. Turn on the current adjusting the rheostat so as to employ a current consumption of approxi- mately 10 amperes and see that the + carbon is the horizontal one. Later when in the prosecution of studies raise the current to one of 15 or even 20 amperes. Move the lens Ci backwards and forwards, at the same time holding a piece of dull black glass, dull black paper, or a piece of ground glass in front of the slit, until a position is obtained which projects an image of the arc of maximum brightness upon the black screen and of such a size as to completely fill the slit opening. Set the slit so that the micrometer screw G is up as shown in the figure, and adjust the opening to about i millimeter by 1.5 millimeters, its length being horizontal. Now hold a black screen against the end of T and move C% back and forth until a very bright and sharp image of the slit is obtained, adjusting Ci again slightly if necessary. Next hold the black screen so that its surface lies in the plane of the optic axis of the observing microscope and adjust the objective 3 so that a very bright, uniform spot of light a little less than i millimeter in diameter is obtained. Turn the fine adjustment V% until the spot of light falls in the optic axis of M. For the final adjustment of the apparatus the cell may be filled with a liquid which contains colloids yielding brilliant diffraction patterns or with a slightly alkaline solution of fluorescein. The path of the illuminating beam is thus easily seen. Focus upon it, using a low power objective and No. i eyepiece and by means of Vi and 2 adjust the beam so that it passes through the center of the field as a narrow thread of light with its minimum diameter at the center of the field. Replace the material employed for ad- justing by the substance to be studied. The only adjustment which should now be required will be the diameter of the slit; if there appears to be required a marked change in slit diameter it is probable that following this change there may be required slight changes of Vi and Vz. If the beginner will proceed as indicated little difficulty will be experienced in adjusting the slit ultramicroscope for use. The most annoying feature is the change in the position of the ULTRAMICROSCOPES 65 crater of the electric arc, and consequently unequal illumina- tion of the slit results or there is a failure (due to a nickering arc) of the spot of light to remain centered upon the slit. Holding the black screen against the lens C 2 , on the side toward the slit, from time to time, will show when the arc needs adjusting, since there should appear a spot of light of uniform intensity and in the proper position to fall concentric with the optic axes of Ci, C 2 , C 8 . When dealing with exceedingly fine colloidal particles it is often an advantage to cut off the lower half of the beam by means of a screen mounted upon a saddle stand and placed between S and C 2 , the upper horizontal edge of the screen being raised so as to cut off the lower half of the beam of light. Ap- proximately as good results may be obtained more easily by laying against the end of the tube T a small rectangular piece of black hard rubber or blackened brass d, as shown in the diagram. Reflecting Condenser Ultramicroscopes consist of highly perfected dark-ground illuminators applied to ordinary micro- scopes provided with special objectives of low numerical aperture. In the special condensers used, the light rays are reflected from two spherical surfaces. The illuminating rays therefore enter the preparations with obliquities greater than in ordinary dark- ground illuminators and are brought to a correct focus. 1 By employing objectives of low numerical aperture (about 0.85) we have rays including only a low range of apertures taking part in the formation of images, although the illuminating rays include a range of high aperture, i.i to 1.35. There is thus obtained the greatest brilliancy of image upon the darkest of backgrounds. Although many different ultracondensers are obtainable, space forbids a consideration of more than two types: the car- dioid condenser of Siedentopf as made by Zeiss, and the ultra- condenser of Jentzsch as made by Leitz. 1 In the ordinary paraboloid condensers, when properly constructed, the light rays are also brought to a focus, but the focal length varies from zone to zone, hence we have an overlapping of images at the center. (Zeiss, "Mikro" Circular 306, p. 8.) 66 ELEMENTARY CHEMICAL MICROSCOPY ULTRAMICROSCOPES 67 The Cardioid Ultramicroscope consists of an ordinary com- pound microscope M, Fig. 30, into whose substage ring the cardioid condenser C is introduced and held in place by the clamping screw /. A thin film of the liquid to be examined is contained in a special quartz cell Q which in turn is held in posi- tion upon the stage in a cylindrical brass mounting B. This mounting may be leveled or slightly adjusted in height with respect to the condenser by means of the screws S. The objec- tive O of the microscope must be specially corrected for use with the quartz cell cover and must have a numerical aperture of less than 0.9. This latter requirement is accomplished by introducing into the objective a funnel diaphragm. As set up for use, the cardioid condenser receives substantially parallel rays from the microscope mirror m. The source of these rays must be some powerful radiant, most conveniently an arc lamp R. Parallel rays are obtained by means of a plano-convex lens L mounted by means of short brass bars r, r, three in number,, attached to the metal screen E. A glass cell W filled with water acts as a cooling trough. A black cardboard or metal diaphragm D serves to cut down the light beam to the proper size for just filling the aperture of the condenser. For convenience in ad- justment as to distance and height, microscope, cell and lens Q. o FIG. 31. Cell for holding Liquids for Study with the Cardioid Ultramicroscope. are placed upon adjustable stands with saddle base resting upon an optical bench of triangular section. The screen E is tipped at such an angle as to project the rays from R upon the properly inclined mirror m, when the latter is at a distance of approxi- mately 60 centimeters from the lens L. The crater of R should be about 8 centimeters from L. The liquid to be studied is placed in a quartz cell Q, Fig. 31, consisting of a grooved quartz disk and cover. With the cover 68 ELEMENTARY CHEMICAL MICROSCOPY in place the liquid forms a thin film q, the excess of liquid being forced into the groove o. The quartz cell is held in position upon the stage of the microscope by means of a brass chamber B consisting of a bed-piece into which the cell fits, a funnel shaped section / pressing gently upon the quartz cover, and a top sec- tion t screwing down upon the section /. When much work is to be done with this device it is best to have all the screw threads but three turns cut from the bed-piece and a slight recess cut as shown at i. This permits a rapid removal of t, and/ is easily lifted out. As furnished by the makers/ is flush with the threads of the bed-piece b and being smooth with no milling is hard to remove. A small pin in/ fitting into a hole in 6, not shown in the diagram, prevents / from turning when t is being screwed down. It is absolutely essential that both cell and cover be absolutely clean and free from all dust particles. Unless so clean that when the cover is laid upon the cell and very gently pressed Newton's rings can be seen the device is unfit for use. To pre- pare Q for use wash very thoroughly both pieces, immerse in hot chromic-sulphuric cleaning mixture, rinse with distilled water, follow by purified alcohol, and dry in a current of warm air, next support upon a loop of platinum wire and heat to a bright red in a Bunsen burner. As soon as the pieces are cool, lay in place in B, and use them at once. When employing the quartz cell and cardioid condenser, never use anything but water as immersion fluid between condenser and cell. Use only sufficient liquid to form a thin layer q and not quite fill the groove o. The objective must be centered by means of the adapter A (Fig. 30), so that the bright spot of light formed in Q will fall in the center of the field. Always raise or lower the cardioid condenser so as to ascertain the proper position for the blackest background and brightest diffraction images. See that the beam of light from the radiant falling upon the mirror of the microscope is of sufficient diameter to fill the aper- ture of the condenser. Use an arc of not less than 15 amperes. ULTRAM1CROSCOPES 69 In the absence of an arc lamp use a 4oo-watt Mazda lamp with concentrated filament. Or if gas alone is available, employ an inverted Welsbach incandescent mantle or even better an acetylene light. Be sure that the reflecting condenser is high enough in its mounting to just touch the object cell upon the stage. Substage ultracondensers are usually screwed into their tubular mountings and are easily turned up or down to permit of their accurate adjustment. The cardioid ultramicroscope is restricted to the study of liquids, to the search for bacteria not readily demonstrated by the paraboloid condenser and to the examination of thin textile fibers, and such other thin semitransparent and flexible solid fragments as will permit pressing out flat, and whose thickness will then be no greater than the thin liquid film of the medium in which they are immersed. Cotton and Mouton's Ultramicroscope 1 consists of a special prism consisting of a rectangular prism of glass having an in- clined face. This prism is laid upon the stage of the microscope and serves for the projection of an oblique beam of light into the preparation placed upon its upper surface. The diagram, Fig. 32, will make clear the construction and the method of using. The prism P, 8 to 10 millimeters high, which converts an ordinary compound microscope into an in- strument for the study of ultra- microscopic particles, rests upon the stage S. The liquid L, to be studied, is placed upon an ordinary glass object slide 5 and covered with a thin cover glass c. A drop of homogeneous im- mersion oil is placed upon the top of P, and the preparation is 1 Cotton et Mouton, C.r., 136 (1903), 1657; Les Ultramicroscopes, Paris, 1906; J. Roy. Micro. Soc., 1903, 573; Lemanissier, Corps Ultramicroscopiques, These, Paris, 1905, 21. FIG. 32. The Ultramicroscope of Cotton & Mouton. 70 ELEMENTARY CHEMICAL MICROSCOPY carefully laid thereon, avoiding all dust particles and air bubbles. This thin film of oil O brings about an optical homogeneity be- tween prism and slide. By means of a condensing lens C of about 15 centimeters focus the rays RRR emitted from an arc lamp as radiant are projected into the prism through the in- clined face, the inclination of this face being approximately 51 degrees. These rays are totally reflected and are brought to a focus at the upper surface of the glass cover at the angle of total reflection. Any particles in suspension in the liquid will diffract the light and diffraction disk images will be formed in the microscope. No other light can enter the instrument and we therefore have the theoretical conditions necessary for the demonstration of ultramicroscopic particles, namely, the par- ticles become luminous upon a black background, the illuminat- ing rays being of high aperture while the image-forming rays are of low aperture. The adjusting of the illumination in this device consists in ascertaining (a) the proper inclination of the rays entering the prism, and (b) the correct distance of C from P, so that the focal point will fall in the proper plane. This adjustment requires considerable care and should first be undertaken by means of some preparation of a colloidal metal (silver, for example), and after having obtained the optimum conditions in this manner, the preparations to be studied are then substituted for the test object. This type of ultramicroscope is applicable only to the examina- tion of liquids. With proper care in adjustment it will yield results fairly comparable with the slit ultramicroscope. In many types of investigation this device possesses a very desirable feature, namely, that of permitting at any time an examination of the preparation by ordinary transmitted light, for it is merely necessary to tip the mirror of the micro- scope and thus send rays M through the object in the usual manner. Absolutely clean glass surfaces free from scratches and inclu- sions are essential. For cover glasses the use of thin freshly- prepared cleavage films of clear mica is suggested by Cotton. ULTRAMICROSCOPES The Jentzsch Ultracondenser 1 can be placed upon the stage of any compound microscope and is so constructed as to combine in itself a reflecting condenser and cell for containing liquids, vapors or gases. It consists, Fig. 33, of a metal cell M, in which are mounted the two reflecting glass bodies G, G'. These are held T/ in place by the cement S, S. Light rays enter the apparatus through the annular opening O, strike the silvered spherical surface in G, are reflected to the curved sides of G' and enter the central cell C. The illuminating rays, therefore, are substantially at right angles to the optic axis of the microscope, thus conforming in general to those in the slit ultramicroscope with, how- ever, this difference, that in the slit instrument the rays enter the cell from, one side only, while in the Jentzsch cell the rays enter from all sides and meet at the center. FlG - 33- The Jentzsch Ultracon- . denser. This instrument may therefore be considered as occupying an intermediate position between the slit ultramicroscope and the cardioid type of ultramicroscope. A cover N fits into the mounting M and is secured in place by a bayonet catch. By turning the cover slightly it is made to press down upon the rubber gasket RR, making a very tight seal against the upper surface of G'. The tubes TT serve for the passage of gas or of liquid through the cell. The cover N is provided with a well-like depression closed at the end by the quartz plate Q. This well permits an objective of long working distance to be focused upon the particles in suspension at the focal point of the illuminating rays. When in use the ultracondenser is laid upon the stage of the microscope with the short tube A inserted into the stage opening. 1 Made by Ernst Leitz, Wetzlar. 72 ELEMENTARY CHEMICAL MICROSCOPY The Abbe condenser is removed or swung aside. The plane mirror is then turned so as to reflect a beam of parallel rays into the device. This beam must be of such diameter as to completely fill the aperture of the condenser. A powerful source of light is essential, preferably an arc lamp or concentrated filament Mazda bulb. The mirror is tipped until the bright spot of light appears at the center of the cell. Since in this case we are examining the path of the rays as in the slit ultramicroscope and these rays enter from all sides and meet at the center, it is unnecessary to exactly center the condenser. Special objectives of great penetrating power are necessary, corrected for the thickness of the quartz plate Q and whose mountings are of sufficiently small diameter to permit their entrance into the well in the cover to a depth such that the focal point will lie within the path of the rays. High magnifications must be obtained by employing high power eyepieces. It follows that there is always an illuminated plane lying below the focal plane of an objective and a perfectly black background is unobtainable. In order to obtain sharper contrasts, a dia- phragm can be placed just above the mirror, either cutting off one side of the beam of light or having an opening slightly eccen- tric to that of the annular opening in the ultramicroscope. Great care must be exercised in cleaning the cell walls and the quartz plate. For coarse colloids and for suspended matter in vapors and water the author has found this device of great convenience and a time and labor saver; but for very fine suspensions the results are not so good. The Immersion Ultramicroscope. In this instrument de- vised by Zsigmondy 1 we have the most improved type of micro- scope for the study of ultramicroscopic particles yet devised; through the employment of immersion objectives of high numer- ical aperture for both illumination and observations, much more brilliant and sharper diffraction disks are obtainable. Thus the existence may be demonstrated of particles even smaller than those rendered visible by ultramicroscopes of the cardioid type. 1 Zsigmondy, Physik. Zeit., 14 (1913), 975. ULTRAMICROSCOPES 73 In this new Immersion Ultramicroscope 1 both the illuminating and observing objectives are beveled at the ends so as to allow their front lenses to be brought very close together with their axes at right angles; the drop of liquid to be examined is placed between the front lenses, clinging by capillarity. No cell is employed. The light rays having but a very short dis- tance to travel, even dark colored liquids may be studied. Diffi- cultly cleanable, expensive cells are thus wholly eliminated, the amount of material required for study reduced to a minimum, and the images obtained are exceptionally brilliant. For the study of hydrosols, water immersion objectives must be used, but for colored glass and similar bodies homogeneous immersion objectives are required. The construction of the instrument is shown in the diagram, Fig. 33A. Fitted to the body tube of a compound microscope is the objective carrier C into which slides a plate to which is screwed the image-forming objective O. To the stage of the instrument is attached the mechanism supporting the illuminat- ing objective I. The micrometer screws S 1 , S 2 , S 3 provide means for the exact adjustment of the beam of light passing in the line of the axis of the objective I, so that it will fall normal to the optic axis of the microscope. S 1 gives an up and down adjust- ment, S 2 forward and back and S 3 from side to side. By rack and pinion S 4 , the entire illuminating device can be lowered for cleaning, for the removal of the objectives, etc. When raised in position for use, the screw s is turned, thus locking the mecha- nism in place. The trough T serves to catch any drip when the liquid is being applied between the objectives. When in use, the instrument is placed on a bed plate with saddle stand upon an optical bench of the type shown in Figs. 24 and 30. An apparatus consisting of a condensing lens and an adjustable slit, also on saddle stands, serves to throw a beam of light from a radiant (arc or Nernst lamp) into the objective I. In critical work the ocular of the microscope is furnished with 1 Made by C. Winkel, Gottingen, Germany. 74 ELEMENTARY CHEMICAL MICROSCOPY an adjustable slit-diaphragm, thus permitting the cutting down of the field until only a certain selected portion is visible. The mutual arrangement of the two objectives is shown in FIG. 33 A. Zsigmondy Immersion Ultramicroscope. the diagram. These objectives embody several unique ideas in mounting, construction and in the component lenses themselves; the end, or front, lenses are of quartz. An examination of the diagram will show that a drop of liquid brought into contact ULTRAMICROSCOPES 75 with the two front lenses will cling in place. The illuminating beam will pass through this drop in the focal plane of the objec- tive O. The image resulting upon focusing the microscope will appear to be two hazy triangles of light united at their apices by a more or less marked brighter thread or band. In this band are seen the diffraction disks due to the infinitely small particles in suspension. By means of the ocular diaphragm all of the hazy triangles are cut off and the connecting thread or band of light alone allowed to appear in the field of view. CHAPTER V. THE EXAMINATION OF OPAQUE OBJECTS, VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES. The study of opaque objects with ordinary compound micro- scopes requires that the illuminating rays shall fall upon the preparations from a point situated above the stage of the instru- ment. This may be accomplished in several ways: (i) the rays from a radiant can be projected upon the surface of the object by means of mirrors, or by means of a condensing lens; (2) a plate of glass or a right-angled prism may be placed above the objective in a tubular mounting so as to fall in the line of the optic axis, and so inclined that any light rays striking the reflect- ing surface will be directed down through the objective, thus brightly illuminating the object. The devices of Group i illuminate the preparation with oblique rays only; those of Group 2 reflect rays perpendicular to the surface of the object and are usually termed vertical illuminators. In the vast majority of cases the examination of opaque objects through illumination by rays normal to the surface of the preparation is preferable to that by means of oblique rays, since the images obtained are brighter, etched figures are more easily interpreted and the finer striations due to incomplete polishing are far less visible. Moreover, photomicrographs are usually more easily obtained. Formerly parabolic reflectors of silvered glass or metal attached to the objective were much employed; but inasmuch as such devices can be used with only a very narrow range of objectives, and with preparations of a certain size only, their usefulness is so limited that chemists have quite generally abandoned them in favor of vertical illuminators. Vertical Illuminators of simple construction consist of tubular adapters or cells so threaded as to permit screwing their upper end into the lower end of the body tube of the microscope, and 76 VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 77 the insertion of an objective into their lower opening. Mounted in the axis of the adapter, or a little to one side, is a reflecting device which receives light projected upon it through an aperture in the walls of the cell and reflects the rays downward through the objective upon the preparation on the stage. The reflecting device consists of a totally reflecting prism or a thin disk of glass or mica. These reflectors are mounted upon small metal rods passing through the adapters at right angles to the optic axis; a milled head at the end of the rod permits chang- ing the angle of inclination of the reflecting surface. In several types the lateral opening for the incident light is made variable in diameter either by means of an iris diaphragm or a rotating collar provided with openings of different sizes. FIG. 34. Prism Vertical Illuminator. FIG. 35. Disk Vertical Illuminator. A typical prism illuminator is shown diagrammatically in Fig. 34. The reflecting device consists of a totally reflecting prism P so mounted as to permit tipping slightly and thus changing the direction of the reflected ray R. Incident light I is projected upon the prism through the horizontal opening 0. A diaphragm D extending not quite halfway across the aperture of the adapter serves to screen the prism and to prevent interfering reflections from blurring the image formed in the microscope. The construction of a disk illuminator is shown in Fig. 35. The incident rays I, I strike a glass or mica disk G and are reflected 78 ELEMENTARY CHEMICAL MICROSCOPY by it through the objective attached below. The rays I, I enter through a circular opening 0. The size of this opening may be changed by turning the collar C which is provided with circular openings of three different diameters. Adjustment of Vertical Illuminators. When the object to be examined is small and is supported upon a glass object slide it is always advisable to place below the object slide a piece of black paper, card or other dark opaque object, so that no trans- mitted light can enter the objective. The size of the spot of light concentrated upon the preparation should correspond approximately to the area of the preparation made visible in the microscope by the particular objective em- ployed. It is therefore desirable that the diameter of the bundle of rays projected upon the reflecting device shall be adjustable. It is also usually best that these incident rays be nearly parallel. These two requirements are met by interposing between the radiant and the illuminator a suitable lens or series of diaphragms. In the better grades of illuminators, lenses and diaphragms are made an integral part of the apparatus. 1 The source of incident light should be a powerful radiant, as, for example, a small arc lamp, tungsten or Nernst incandescent, or inverted Welsbach gas burner, acetylene light, or stereopticon lamp with concentration filament, or better still a nitrogen filled tungsten. In all cases the radiant should be as close to the illuminator as is possible for convenience and safety. With powerful radiants and condensing lenses, it is wise to interpose between radiant and illuminator a water cell of moderate thick- ness to act as a cooling device. With very highly polished surfaces the image obtained is often of such dazzling brightness as to be almost blinding; in such cases a piece of greenish or blackish glass should always be inter- posed between radiant and illuminator or placed above the eye- piece. \ 1 The 4 to 5 ampere arc lamps for microscopic purposes are generally fitted with a plano-convex condensing lens; in such an event no other lens between radiant and illuminator may be required. The lamp should stand 8 to 12 inches from the illuminator. VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 79 Nernst lamps with very small incandescent filaments often fail to yield a sufficiently even illumination; under such con- ditions a piece of ground glass interposed between lamp and illu- minator will usually greatly improve the field of view, but will of course reduce the brightness of image. To obtain satisfactory results in the study of opaque objects with vertical illuminators it is important that the objectives em- ployed be constructed with compact mounts and that the lenses be corrected for use with uncovered objects. Standard microscope objectives are always corrected for some definite cover glass thickness. Moderate or high power objectives of this sort, therefore, cannot be employed for the study of uncovered preparations. Most objective manufacturers supply special objectives for use with vertical illuminators. Such objectives have very short mounts and have the rear lens combination flush with the upper edge of the mount (see Figs. 36 and 46). This is done to prevent internal reflections and yields better fields and clearer and brighter images. It is a safe rule to follow, if the best results are wanted, to select an outfit in which the distance between the re- flecting surface of the illuminator and the rear lens combination of the objective is as small as possible. The diagrams, Figs. 36 and 39, have been drawn with a view of showing this in an exaggerated way. In Fig, 36 a short com- pact mount is shown, the rear lens combination is almost in contact with the reflecting prism P, while in Fig. 39 an ordinary objective is shown and the distance between reflecting disk F and the rear lens is so excessive as will doubtless lead to interfering reflections of an aggravated sort. With the construction shown in Fig. 39, an objective with compact mount would be essential. The interior walls of vertical illuminators must never be al- lowed to become bright but must be kept coated at all times with a dull black finish. Since the diameter of the rear lens combination is different in different objectives, especially when manufactured by different firms, it is evident that the best results will be obtained with illuminators of the prism type, only when the prism can be dis- 8o ELEMENTARY CHEMICAL MICROSCOPY placed forward and back with reference to the optic axis of the objective in order that just the proper area of the objective may be covered by the prism. When properly adjusted the image of the illuminated prepara- tion should be of uniform intensity throughout and should not have half the field hazy and blurred with a whitish fog. Chang- ing the distances between radiant, collective lens and illuminator and tipping the prism slightly will improve matters, but with illuminators of the type shown in Fig. 34, there sometimes re- mains a slight blurring of half the image. To meet this difficulty, two sliding diaphragms are provided in the Zeiss illuminator, which slip into the slot S, so constructed with two apertures and a central opaque stop as to effectually prevent reflections and passage of rays from the prism in line with the optic axis of the objective. When adjusting the illuminator, first one, then the other, of the two diaphragms should be tried to ascertain which will yield the clearest image, observations being made with each diaphragm inserted to different depths; an exceedingly slight displacement very seriously affects the clearness of the image. Interpretation of Appearances with Vertical Illuminators. The investigator is generally dealing with more or less highly polished surfaces and with areas, part of which are polished, part rough and often studded with minute bristling points. Less frequently, as, for example, in the study of material exhibiting fatigue failure, the preparations are polished but are crossed by exceedingly minute cracks or cleavage planes. To ascertain whether the surfaces are polished or mat, whether we have to deal with elevations or with depressions and to enable us to dem- onstrate slip bands in fatigue failure requires that we shall be thoroughly familiar with the optic effects resulting from different types of illumination by reflected light. These effects have al- ready been discussed at length on pages 38 and 39, to which the student is referred. With ordinary etched metal preparations no special difficulties arise, for with vertical illuminators the polished surfaces appear bright, the irregular or mat surfaces more or less dark. But to demonstrate fissures, cleavage planes, depressions, etc., requires VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 8 1 that the examination with the vertical illuminator be supple- mented by very oblique illumination and that due account be taken of the directions of shadows with respect to the radiant, remembering of course that in the image seen in the microscope directions are completely reversed. Polarized Light with Vertical Illuminators. A further aid in differentiating between the phases present in a given specimen is afforded by employing polarized rays for illumination or analyzing the light rays reflected from the object. The light rays reflected from the polished surfaces of sections of aniso- tropic crystals are polarized, as has been already stated, while the rays reflected from isotropic crystal sections are not polarized. It is evident that if we pick out a given phase and employ a magnification, such that an area of this phase alone fills the field, we can, by studying the nature of the light reflected therefrom, often obtain information of the greatest value as to the nature of the composition of the specimen being studied. In many instances it is not even essential to confine one's attention to a selected small area but we may use low powers which will include several phases in the image. FIG. 36. Nachet Vertical Illuminator. Nachet Vertical Illuminator. 1 This instrument, Fig. 36, consists of a collimator tube C attached to a cell F, which in turn slips into the threaded adapter A and is held in place by the thumb-screw B. The adapter A carries at its upper end a male screw thread of standard pitch, serving to fasten the device into the end of the tube T of the microscope, while F is tapped with 1 Manufactured by A. Nachet et Fils, Paris, France. 82 ELEMENTARY CHEMICAL MICROSCOPY standard thread for the attachment of the objective OO'. Lying in the axis of the tube C is the reflecting prism P, the surface R of which is silvered, and the outer end L ground convex, thus serving the purpose of a plano-convex collecting lens. An iris diaphragm whose diameter is adjustable by the knob K is fas- tened eccentrically to C. The position of the center of the dia- phragm with respect to the axis of C may be changed by loosening the screw S, thus making it possible to alter the position of the point of incidence upon R of the illuminating rays from the radiant, according to the power and mounting of the objective employed. The light rays proceeding from the radiant pass through the lens L, and striking the surface R, pass through the objective which now acts as a condenser, throwing a tiny spot of intense light upon the surface of a metal preparation M. The light rays reflected from M reenter the objective to form the image seen in the microscope. A noteworthy feature of this type of vertical illuminator is the placing of the prism P in such a position as to bring its lower surface as close to the upper lens combination of the objective as it is possible to do. This greatly reduces the danger of the formation of a hazy or cloudy image by eliminating internal reflections. The position of the prism P is fixed, hence all adjustments of the light rays must be made by displacing the iris diaphragm and thus changing the position of the spot of light upon the reflecting surface R. The Leitz Vertical Illuminator 1 is so constructed as to permit the insertion of either a disk or a right-angled reflecting prism above the objective, and is therefore applicable to all heights and powers of objectives. The construction is shown in Fig. 37. To a cylindrical adap- ter K a collimator tube T is attached which carries a condensing lens L in its mounting C. C slides within T, thus permitting regulation of the diameter of the illuminating beam of light pro- jected upon the reflecting surface. One side of K is flattened and through this surface is cut an opening into the interior of the cell. The lower part of this opening is dovetailed as shown at d. 1 E. Leitz, Wetzlar, Germany. VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 83 The prism P and the disk k are attached respectively to the axis of the milled wheels W and W. These in turn are mounted upon metal plates with edges obliquely cut so as to fit into the dove- tail d. These plates when inserted and pressed in place are held by the spring s. They are thus se- cured in proper position but can be slid back and forth in the slot d. A mark S upon the plates and an- other / upon the adapter serve to indicate the proper position of P or k with respect to the optic axis of the microscope M. To remove the prism, the wheel W is pressed gently downwards and outwards, thus re- leasing the plate from the spring s; W is then carefully raised until the plate is free from the slot d. It can then be removed by tipping up slightly and withdrawing from the opening. To insert the disk, turn W until the groove i is horizontal, introduce k into the opening and push down till the lower edge fits into d, then press W forward as far as it will go. The groove S is then brought into coincidence with /. The reflecting disk k is fastened to a mounting by the spring fingers v. This device permits the rapid and easy removal of the disk for cleaning or for replacement when broken. The objective O is screwed into the lower opening of K; in the illus- tration is an 8 millimeter apochromatic, for 200 millimeters tube length, uncorrected for cover glasses. Just as in the simple prism or disk illuminators, the rays of light striking the reflecting surface are directed downwards through the objective upon the object m. Parallel light should fall upon the lens L. This is obtained by employing a suitable lens between the illuminator and radiant. The Leitz Company supply a very conveniently mounted lens for this purpose. A metal screen A, Fig. 38, is attached to a stand B. Mounted in the screen is a lens in front of which is an FIG. 37. Leitz Vertical Illuminator. ELEMENTARY CHEMICAL MICROSCOPY iris diaphragm D. The stand and radiant are placed at such distances from L as to project a small beam of approximately parallel light upon L. The milled head a serves as a fine adjust- ment up and down of the lens and diaphragm. When either daylight illumination, direct sunlight, or a radiant at a distance are to be used, the mirrors R 2 and RI are brought into service, the light from the chosen source being received upon R 2 , reflected upon RI, and thence through the lens and dia- phragm opening. When a radiant close to A is used the mirror RI is raised until it stands in a vertical position, thus giv- ing an unobstructed passage through the center of A. Correct illumination of the surface of an object m is obtained as described above by trying the lens L at different distances from P and by tipping P or k FIG. al^ndensing L 7ns until the most satisfactory angle of in- and Iris Diaphragm for clination is obtained. It may also be Use with Leitz Vertical necessary to slide S slightly to the right Illuminator. Qr Ht Q ^ indicator t It is usua lly best to start with a diaphragm opening yielding a beam of light which will not more than half fill the aperture of the lens L. Tassin Vertical Illuminator. One of the greatest annoyances encountered in the work with ordinary vertical illuminators is the necessity of readjusting the height of the radiant whenever a change of objective is made or objects of different thicknesses are studied, since refocusing is essential and this necessarily alters the position of the disk or prism with reference to the axis of the radiant. To obviate this defect Tassin has devised an appara- tus in which the radiant either a small tungsten lamp or an acetylene burner is attached to the illuminator mounting and hence in focusing, both radiant and illuminator are displaced simultaneously an equal amount; thus no realignment is neces- sary. The construction of this device will readily be understood VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 85 by referring to the diagram, Fig. 39. An ordinary disk (or prism) illuminator I is attached to the tube T of the microscope. Into the lower opening is screwed an aluminum adapter A which serves to hold in position the supporting bar B. The objective O is screwed into the lower end of A. The bar B carries a vertical sleeve J, fitted with a thumb-screw and serving to hold in place FIG. 39. Tassin Vertical Illuminator. the remaining parts of the illuminator. The sleeve D carries a Ramsden eyepiece, securely held in position by the screw K. This eyepiece acts as a condensing lens. The correct position of the lenses to obtain a spot of bright light of the requisite diameter upon the reflecting surfaces is secured by sliding the entire ocular in the -sleeve or by sliding the lens C or both. The clamping joint E permits tilting the condenser so as to obtain the correct angle of incidence upon the disk or prism. To ex- clude all other light from the illuminator, a screen S is attached 86 ELEMENTARY CHEMICAL MICROSCOPY to the condenser system. Fastened to S is an arm G which carries the radiant R. In the diagram the radiant is an acety- lene light, adjustable both up and down and forward and back in the mounting H. To make the nature of the burner clearer the flame is shown with its broad side toward the condenser. This is, however, an incorrect position for use, the proper posi- tion being always with the edge of the flame toward the illuminator in order that the full intensity of the radiant may be obtained. When, instead of the acetylene burner, a tiny tungsten lamp is supplied for use with this device a parabolic cover and reflector is placed back of the bulb and holds it in proper place against the screen (see Fig. 48, page 99). The light rays from the radiant pass through the condenser system, strike the reflecting device of the illuminator and are totally reflected down through the objec- tive O upon the specimen M. The light rays reflected from M pass through the objective and strike the disk F at an angle other than that of total reflection and thus pass through to form the image in the ocular of the microscope. Owing to the relatively great distance between the reflecting disk F and the objective it is essential that the inner surfaces of I, A and O be kept a dull, black in order to prevent internal reflections. The disadvantage of employing ordinary objectives instead of those in special short mounts will be apparent at once from the diagram, for, as just pointed out, the danger of internal reflec- tions is very great; moreover, the length of I and A prevent low powers from being employed unless the microscope is provided with a substage upon which the specimen can be supported. With specimens placed upon the stage any attempt to focus the upper surface will entail raising the body tube of the microscope until the rack and pinion are out of mesh. A very simple and efficient device is shown in Fig. 40, in which the radiant moves with the illuminator in focusing, thus avoid- ing the necessity of realignment for different- sized specimens or changes of objective. It consists of a thin, bent aluminum plate S inserted between the body tube T of the microscope and the vertical illuminator I. Being provided with a hole a trifle larger VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 87 than the size of the illuminator threads it may be moved to the right or left for alignment, and clamped fast by turning the screw collar of I. L is a 6-volt i6-candle-power tungsten lamp coated with "frosting compound" in which fine graphite has been sus- pended. This provides an absolutely opaque covering for the lamp and prevents annoying side lights in working. A tiny clear area is made in the coating at O by means of a little alcohol FIG. 40. Vertical Illuminator with Simple Device for a Tungsten Lamp Radiant. on a bit of rag or cotton. The lamp L is attached to the end of S in a slot cut for the purpose and is secured in place by wires soldered to the terminals of the lamp. To obtain rigidity these wires pass downward through bits of glass tubing i as insulators. S is so bent as to bring the glowing lamp filament in line with the center of the illuminator diaphragms. As shown in the illus- tration the whole device, including the coating of the lamp, can be made in any workshop in about an hour. A better arrange- ment, when shop facilities permit, is to fasten an attaching socket with bayonet catch to S. The lamp is connected with the usual no- volt lighting circuit with the interposition of a suitable rheostat or lamp bank, allowing the passage of 2 to 3 amperes (five i6-candle carbon filament lamps or two 32 and one 88 ELEMENTARY CHEMICAL MICROSCOPY 16). Instead of the i6-candle-power lamp one of about 20 candle- power may be substituted. The author has found this simple and inexpensive apparatus very satisfactory. A clear, bright, uniformly illuminated field is obtained and there are no adjustments necessary. FIG. 41. Chemical Microscope with Stage removed and Auxiliary Stage inserted in the Substage Ring. Maintaining the Alignment of Radiant and Illuminator may readily be accomplished in microscopes provided with an adjust- able substage by removing the condenser or polarizer and sup- porting the specimen upon the substage ring. In the case of the chemical microscope, the stage is removed by loosening the VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 89 centering screws and lifting out the stage. An "auxiliary" stage is then inserted in the substage ring, the specimen placed upon it and the focusing is done by means of the substage quick- acting screw. Delicate focusing may then be made by the fine adjustment of the microscope. This method possesses the ad- vantage of producing no disturbance of the alignment of radiant and reflector in changing objectives or in studying successively preparations of greatly varying thickness. Fig. 41 illustrates the chemical microscope with auxiliary stage applied for the examination of opaque objects. The auxiliary stage itself is shown at A. Mounting Polished Objects. In order to mount small preparations for examination with vertical illuminators so that when placed upon the stage of the microscope, the upper or pol- ished surface will lie in a plane at right angles to the optic axis of the microscope, proceed as follows: place upon a i by i \ inch extra thick object slide of metal or glass a small piece of soft plas- ticine, soft beeswax or soft paraffin; lay the object to be studied polished side up upon the imbedding material and place the prepa- ration upon the substage ring (with auxiliary stage in place if one is at hand) ; place a thick ~ '. , r FIG. 42. Device for glass object slide upon the stage of the micro- Mounting Pieces of scope and then carefully raise the preparation Polished Metal for by means of the substage screw until it is Study with Vertical - . , . i.,. Illuminators. pressed firmly against the object slide, the latter being held in place with the fingers. The upper surface of the object to be studied is thus made parallel to the plane of the stage and is in proper position for examination with the vertical illuminator. Special mounting cells employing this same principle have been designed. One of these cells or devices is shown in Fig. 42. It consists of a bed plate attached to a base and threaded to carry a collar screwing up and down. The upper edge of the collar is exactly parallel with the surface of the bed plate. The collar is screwed up or down to accommodate specimens of different thicknesses. go ELEMENTARY CHEMICAL MICROSCOPY The specimen to be mounted is laid upon a piece of lens paper, polished side down upon the bed piece. The collar is then raised or lowered the proper amount and an object slip carrying a bit of plasticine is inverted over the preparation and pressed down until each end touches the circumference of the collar. The slip may now be lifted off, carrying with it the specimen imbedded in the plasticine or wax. Laid upon the stage of the microscope, the polished surface of the specimen will be in a plane normal to the optic axis of the microscope. Metallurgical Microscopes. The extraordinary interest in the microscopic study of metals and alloys within the last ten years and the astonishing development of theories relative to their constitution and structure, followed by the application of this information to the mechanic arts, has led to the design of special forms of microscopes to facilitate the study of the many different problems arising in the metallurgical industries. In all these special types of microscopes we have to deal with compound microscopes, having permanently attached, between ocular and objective, a vertical illuminator, usually of the prism type. Since the etched surfaces of metals ordinarily yield images of such intricacy that notebook sketches become impracticable, recourse must be had to photography. Most metallurgical microscopes therefore include as an integral part of the in- strument a photographic camera, and when thus provided they are often known as metallographic microscopes or metallo- graphs. In order that the structure of an alloy may be studied it is essential: (i) that a small area shall be ground to a plane surface, polished and etched; (2) that this plane surface shall lie normal to the optic axis of the microscope; (3) that the area of this plane shall be so situated with reference to surrounding parts that the objective may be brought sufficiently close to it to be focused. Were the preparation to be laid upon the stage of an ordinary microscope it would have to be thin and to have another sur- face ground parallel to the etched surface. To avoid these VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 91 difficulties and further to permit the examination of fragments of moderate size, the microscope is more conveniently inverted, i.e., constructed with the objective lying below the stage. The alloy can thus be laid upon the stage, polished surface down over the stage opening. It will thus meet the requirement that its etched surface shall lie in a plane normal to the optic axis. Coarse adjustment focusing is accomplished by displacing the stage up or down, the tube of the microscope remaining in a fixed position, assuring no disarrangement of the proper align- ment of the illuminator with reference to the radiant. Most of the large metallographs are developments of the type first suggested by Le Chatelier. Two instruments have been selected for illustration as embodying the largest number of good features to the exclusion of those which are distinctly bad. These have been described at length in preference to other valuable instruments since the author has had the opportunity of working with them and thoroughly testing them. The Leitz Metallurgical Microscope. This instrument con- sists of the vertical illuminator shown in Fig. 37 applied to a compound microscope so arranged as to lie in a horizontal posi- tion. The general arrangement of its component parts is illus- trated by Fig. 43. An optical bench B carries a series of stands with saddle bases. These stands support the different unit parts of the instrument. The first stand carries a small arc lamp R, a condenser C and a screen E ; attached to E is an iris diaphragm D and a shutter (not shown in the cut) for making the photographic exposures. The next stand supports the com- pound microscope and its accessories; the mechanical stage S/, the illuminator I fitted with iris diaphragm d, the reflecting prism V and the body tube of the microscope M with its ocular; over the ocular is fitted a removable black glass disk b. Coarse adjust- ment of the microscope is obtained by the wheel F which raises and lowers the stage St supported by four pillars; the object to be examined is placed polished side down over the opening of the stage; the rays of light projected by the radiant enter the illumi- nator I, are then reflected upward through the objective and strike the surface of the object; the rays are thence reflected ELEMENTARY CHEMICAL MICROSCOPY VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 93 downward through the objective into the prism V, the inclined surface of which reflects the rays into a second right-angled re- flecting prism attached to the end of the tube M. This latter prism sends the rays to the eye through the ocular of the micro- scope in the upper end of M. The entire tube M slides into a FIG. 44. Leitz Metallurgical Microscope fitted with Abbe Condenser and Mirror. sleeve: by pulling out M, the prism attached to the lower end is withdrawn from the path of the rays which then pass horizon- tally through the tube N carrying at its end a projection eyepiece. The image is thus formed upon the ground glass or photographic plate of the camera. A pair of Hooke's keys K, K serve to focus the image upon the ground glass. The rod r rising from the stage serves to attach a mirror and an Abbe condenser. By 94 ELEMENTARY CHEMICAL MICROSCOPY removing the illuminator, inserting an objective directly over the prism V, and attaching an Abbe condenser A and mirror m to the rod r, the microscope may be employed for the examina- tion of transparent objects by transmitted light. This arrange- ment of the instrument is shown in Fig. 44. The adjustment of the illumination in the Leitz metallurgical microscope is in every way similar to that followed in the Leitz vertical illuminator already described. For high powers the makers suggest employing only the disk reflector, for moderate powers the reflecting prism is used, while for very low magnifica- tions a plate reflector is supplied so arranged as to fit between the objective and the preparation. This last device is restricted to such low magnifications, however, as to be rarely applicable to ordinary metallographic studies. The Reichert Metallurgical Microscope. This instrument, Fig. 45, is one of the most convenient and most substantially built of its class. The microscope itself consists of a heavy base A from which rise four pillars; the largest of these, fashioned into a handle H, carries the stage S provided with rack and pinion adjustment for roughly focusing the preparation. The pillar P supports the microscope proper, consisting of a prism chamber to which the objective O, the illuminator tube V and the body tube are attached. The fine adjustment of the instrument is accomplished by the milled screw F. The prism chamber fur- ther carries a tube whose axis is at right angles to that of T, fitting into a light tight sleeve joint in C. The tube C, sup- ported by a pillar provided with rack and pinion vertical adjust- ment, fits into the front of the photographic camera and serves as a carrier for a projection eyepiece. To the fourth pillar is attached the tube B, fitted with a lens for projecting parallel rays into V, a rotary disk provided with diaphragm openings of different sizes, a glass cooling cell W for water or for colored solutions to be employed as color screens and a slot for the insertion of green, yellow or black glass slips G to modify the brilliancy of the strongly illuminated surfaces. For greater con- venience in making photographs a shutter s for exposures may conveniently be attached to the end of the tube B. VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 95 9 6 ELEMENTARY CHEMICAL MICROSCOPY In the Reichert instrument the illuminating and image-forming rays take the directions indicated in Fig. 46. The light rays from a radiant R enter the tube V, pass through a condensing lens and are twice reflected by the prism P. Passing through the objective O, a spot of bright light is formed upon the surface of the inverted preparation M lying upon the stage S. The surface of M reflects the rays downward through O to the prism PI, thence the image-forming rays are reflected in the direction E FIG. 46. Path of Light Rays in the Reichert Metallurgical Microscope. through the body tube T of the microscope which carries at its outer end an eyepiece of the usual construction. Owing to the space required for mounting PI, the tube length of the micro- scope is greater than usual and objectives corrected for a tube length of not less than 200 millimeters must be employed. The prism PI is so mounted that it can be rotated through an arc of 90 degrees by means of the milled head K. To K is at- tached an indicator I which marks the position of the reflecting surface. In the position of I shown in the diagram the image formed by the objective is sent to the observing tube of the microscope; turned through 90 degrees the rays are reflected into the tube indicated by the dotted circle C, whose axis is at right angles to that of T. The tube C is fitted with a projection ocular whose function it is to form an image upon the ground VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 97 glass or photographic plate of the camera. Set screws H H serve to adjust and to limit the amount of rotation of K so that when turned as far as it will go in either direction the images formed will be comprised in circular fields uniformly illuminated through- out their entire areas. The illuminator prism P is so mounted as to permit its hori- zontal displacement below the objective in order to conform to the requirements of different objectives. Were P immovable, in the majority of cases half the image formed in the eyepiece would be covered with a whitish cloud or fog. The amount of horizontal displacement is also such that P can be slid entirely beyond the aperture of the objective, thus allowing an unob- structed passage of rays from the objective to PI. Transparent objects placed upon the stage can thus be studied by transmitted axial light or by oblique light. To prevent undue strains upon the stage when large specimens must be examined a supporting rod L rises from the pillar to which C is attached and passes through a clamp attached to the pillar support of C. Tightening a set screw in the clamp pre- vents any vertical movement on that side of the stage and as a further protection, a set screw is also provided for locking the coarse adjustment in position. The stage may thus be held rigidly in any plane and relatively heavy objects may safely be laid upon it. The mirror m is attached to a bar of such length that it can be swung close to the objective and serves to project oblique light upon the surface of the preparation in the study of fissures, slip bands, cavities, etc. The radiant R consists of an arc lamp, that shown in the figure being a "hand feed" type. The body of the lamp is raised upon its supporting rod until the center of the crater of the arc lies in the line of the optic axis of the lenses mounted in B and V. Hooke's keys Ni, N2, NS permit adjustment of alignment and illumination while looking into the instrument, NI turning the lamp from side to side, N 3 up or down, and N 2 approaching or drawing apart the carbons. In this lamp both -f- and car- bons are of the same size, but the pitch of the screw threads 98 ELEMENTARY CHEMICAL MICROSCOPY feeding and moving them are in the ratio 2:1. Hence the move- ment through the key N 2 compensates for the more rapid con- sumption of the horizontal carbon. The projection eyepiece in C forms an image upon the ground glass or photographic plate of the camera U, the size of the photo- graph taken being regulated by extending or contracting the bellows of the camera. A graduated scale engraved upon the optical bench, upon which the camera slides, permits a record being kept of the position of the photographic plate at the time of the exposure. In order to obviate the necessity of passing to the rear of the camera to look at the ground glass and to focus, a mirror is placed within the camera box and hinged to one cor- ner; by means of the lever I this mirror may be swung diagonally across the box at an angle of 45 degrees. The image of the preparation will thus be projected upon this mirror whence it is reflected upon the glass g. The investigator can thus examine the image and focus the same without leaving his seat before the body tube T. When an exposure is to be made, the lever / is pushed back until the mirror lies flat against the side of the camera. There is thus an unobstructed passage through C and the camera to the photographic plate. Although the Reichert metallurgical microscope is one of the best and most convenient of its kind, it has at least one very serious defect. It is supplied with specially constructed objec- tives whose mounts are of small diameter so made as to drop into a sleeve or adapter instead of screwing in as is the case with ordinary objectives. The purchaser of the instrument must therefore obtain his entire outfit of objectives at the time the microscope is bought. If future purchases of objectives are required it is necessary to obtain them from Reichert and in order to be certain of their proper centering the microscope should be sent to factory. 1 1 This difficulty has been eliminated in the instrument in the author's laboratory by fitting the opening above the illuminating prism with standard or international thread ("society screw"), thus permitting the use of all standard objectives irre- spective of the firm manufacturing them, and thus greatly increasing the usefulness of the microscope, particularly when the instrument is employed for the examina tion of transparent objects by transmitted light. VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES 99 Metallurgical Microscopes for the Examination of Large Castings, etc., are now manufactured by a number of different firms. Such instruments are often designated, as "Works Microscopes/' since their purpose is the study of materials of construction already in place or too large to bring into the laboratory. FIG. 47. Stead Works Micro- scope. FIG. 48. Tassin Metallurgical Microscope. As indicated by the name and purpose they are compact, sub- stantially built and easily transportable. They consist essen- tially of a compound microscope, whose pillar or handle arm has been separated from the remainder of the instrument in a line in the plane of the stage, and attached to a suitable base or to three legs. In other words, these instruments are microscopes without stage or substage. When in use, the base rests upon the object to be studied and the tube carrying objective, illumi- nator and ocular is raked down until the surface of the object is in focus, there being an aperture in the base in line with the optic axis or the base is provided with widely divergent legs. Figs. 47, 48 and 49 illustrate typical instruments of this class. 100 ELEMENTARY CHEMICAL MICROSCOPY In the Stead instrument, Fig. 47, the body tube is supported upon three adjustable legs. Focusing is done by hand by rais- ing or lowering the tube in a sleeve. When in focus the instru- ment is held in place by a clamping screw C. A vertical illumi- nator of the disk type forms an integral part of the instrument. 1 The radiant in this case consists of a tiny incandescent electric lamp enclosed in a sleeve at right angles to the illuminator mount- ing. As the instrument is intended for low magnifications only, no fine adjustment is provided. A somewhat similar idea in illum- inator construction is found in the Tassin metallurgical microscope. 2 In this instrument, Fig. 48, we find the illuminator of the form already FIG. 49. Leitz Metallurgical Microscope. described on page 84, Fig. 39, the radiant being either an elec- tric or an acetylene lamp. The microscope itself has no substage but is mounted upon a heavy base with central opening and pro- vided with four large leveling screws. The third type of instrument is illustrated by the Leitz metal- lurgical microscope, Fig. 49. Here we have a compound micro- 1 See Stead, Work Shop Microscopes. J. Roy. Micro. Soc. 1909, 20, 22. 2 For its application see Tassin, The Microstructure of Steel Castings, J. Ind. Eng. Chem., 5 (1913), 713. Metallography as Applied to Inspection, J. Ind. Eng. Chem., 6 (1914), 95. VERTICAL ILLUMINATORS, METALLURGICAL MICROSCOPES IOI scope, consisting, as usual, of stage and substage, but with this difference, the tube and pillar are detachable from the stage, and the substage and support detachable from the base. By attach- ing the microscope and pillar to the base there is obtained a works microscope applicable to the study of large castings. The area of the casting to be studied is visible in the microscope in the opening between the legs of the horse shoe base. Light from a suitable radiant is deflected by the mirror m into a right-angled prism attached to the end of the illuminator. For the proper illumination of the objects, the methods and precautions already described on pages 78 to 82 are obviously equally applicable. U ; x .if s. CHAPTER VI. USEFUL MICROSCOPE ACCESSORIES, LABORATORY EQUIPMENT, WORK TABLES, RADIANTS. Drawing Cameras (Camera Lucidas). It is very frequently the case that sketches, relative proportions of structural details, or actual measurements of component parts of preparations being studied must be entered into notebooks. Free-hand draw- ing is tedious, difficult, and if a sketch to scale is required, as is usually the case, an exceptionally good judgment of proportion is essential. To obviate these difficulties a drawing camera may be employed. Although there are many types of these devices upon the market, the chemist is usually restricted to those forms which permit employing the micro- scope in a vertical position. The most convenient of these drawing cameras are shown in Figs. 50 and 51. If, after attaching one of these devices to the tube of the micro- scope above the ocular, the worker looks into the instrument, he is able to see simultaneously both the preparation and the page of the notebook. In the forms shown in Figs. 50 and 51, known as Abbe prism 'camera lucidas, there is placed above the ocular a cube of glass which has been cut diagonally, the surface of one-half being silvered and cemented again in place, after a central oval per- foration has been made through the silvered surface. This oval aperture allows the image-forming rays of the microscope to reach the eye while the silvered surface reflects from a mirror the image of the notebook page or drawing paper. Fig. 52 shows diagram- matically the path of the light rays, the dotted lines indicating FIG. 50. Small Abbe Drawing Camera. (Bausch & Lomb Optical Co.) USEFUL MICROSCOPE ACCESSORIES 103 FIG. 51. Large Abbe Drawing Camera. (Spencer Lens Co.) FIG. 52. Diagram of the Path of Light Rays in Abbe Drawing Cameras. 104 ELEMENTARY CHEMICAL MICROSCOPY the image-forming rays from the drawing paper BB reflected by the mirror M to the reflecting surface ef of the Abbe prism P, and thence to the eye of the observer. The solid lines indicate the image-forming rays from the preparation upon the stage of the microscope, passing through the aperture in ef also reaching the eye. It is obvious that the observer is able to see both the image of the preparation and the drawing paper and can there- fore trace upon the paper with a pencil the outlines and many details of structure of the preparation. In order to avoid distortion of the drawing the mirror M must be so inclined that the light ray be shall fall normal to the paper. From an examination of the diagram it will be seen that unless the opening in ef is placed at the eye-point considerable light will be lost and unsatisfactory results will be obtained. Before at- taching a drawing camera always first ascertain the position of the eye-point (see page 13). It not infrequently happens that in designing an ocular, the manufacturer fails to take into account the fact that the investigator may wish to use a drawing camera. The eye-point may in such cases lie so close to the eye lens or may lie so far above it as to render the employment of an Abbe prism camera impracticable. Because of this great difference in the relative position of the eye-point in different oculars it is best, in purchasing an Abbe camera, to select one of the type shown in Fig. 51, since in instruments of this sort the prism mounting is of the smallest dimensions possible and the distance between prism and clamping ring will allow exceedingly great latitude in movements up and down. In order to equalize the light intensity reaching the eye from preparation and drawing paper, a series of dark glasses of graded degrees are mounted so as to turn and be swung in position, by a ring between prism and paper, and a ring between prism and ocular. By properly adjusting the diaphragm of the Abbe con- denser and then selecting the right glasses in these rings, it is always possible to obtain a clear image of both preparation and drawing pencil. The large cameras of the type just referred to, are provided with a graduated extension bar to which the mirror is attached USEFUL MICROSCOPE ACCESSORIES I0 5 to facilitate adjustments, and the axis upon which the mirror tips is graduated into degrees. When the paper lies horizontally with respect to the optic axis of the microscope, the mirror should be set at 45 degrees, providing that the mirror bar is long enough to prevent interferences due to a reflected image of the stage; if not, then the mirror must be tipped to an angle nearer to the horizontal and the drawing paper inclined until the central rays become normal to it. The amount of inclination of the drawing surface must be twice as many degrees as the mirror is tipped below 45. Camera lucidas serve not only for drawing but are most useful in micrometry, in reading thermometers when melting, boiling or subliming points are determined, or in reading scales of small voltmeters or ammeters when observations are being made, for upon looking into the micro- scope both the preparation and the scale of the instrument may be seen. The Leitz Drawing Eyepiece, shown in section in Fig. 53, consists of a neg- ative eyepiece whose lenses are so mounted as to permit the insertion of a reflecting prism P just above the eye lens extending to the optic axis of the ocular. Light rays (as indicated by the dotted line) from the drawing paper enter the prism, are twice totally re- flected from the inclined surfaces of the prism and enter the eye together with the image-forming rays of the microscope. The eye there- fore perceives the image of the object under the microscope apparently projected upon the drawing paper. Neutral tinted glasses N serve to reduce the light intensity from the drawing paper and to thus facilitate following the tracings of the pencil point. The screw S serves to clamp the device in place while in use. Two types of these Drawing Eyepieces are manufactured, FIG. 53. Drawing Eyepiece. (E. Leitz.) 106 ELEMENTARY CHEMICAL MICROSCOPY one for use with the microscope in a vertical position, the other for a slightly inclined instrument. Since the prism forms an integral part of the eyepiece, changes in magnification must be made wholly by changing objectives or changing the distance from drawing board to prism. Microspectroscopes or Spectroscopic Oculars consist of direct vision spectroscopes as integral parts of microscope eyepieces. They are usually constructed after the Sorby-Browning pattern, using a compound direct vision Amici prism. These prisms consist of either three or five units, a prism of flint glass between two of crown glass, or two prisms of flint glass alternating with three of crown glass. This prism is mounted just above the eye lens of the ocular, while the slit of the spectroscope is placed in the plane of the diaphragm of the eyepiece. Usually a com- paring prism is provided, which, when in position, cuts off half the width of the spectrum and permits placing in juxtaposition with the spectrum of the material being studied, the absorption spectrum of a solution of known composition. The position of bands or the amount of the spectrum cut off is determined by an arbitrary scale; or by means of an Angstrom scale reading in wave lengths, projected upon the spectrum, or by means of some indicating device moving the length of the spectrum, its position at any given point being indicated by a scale moved by a microm- eter screw. This last type is the only one of value to the chemist. The microspectroscope illustrated, 1 Figs. 54 and 55, is pro- vided with a measuring device capable of yielding concordant measurements with a very fair degree of accuracy. The instru- ment consists of the cell or chamber K in which are housed the slit 5, the comparing prism p, a movable diaphragm d, and in the lower opening the field lens / of the ocular. A small opening O in the side of K permits light, reflected by the mirror m, to enter the prism p and thus yield a spectrum in juxtaposition to that obtained from the object under the microscope. The solution or transparent solid used for comparison is held before the open- ing O by means of the clamps CC. The knob P serves to swing 1 Manufactured by W. & H. Seibert, Wetzlar, Germany. USEFUL MICROSCOPE ACCESSORIES 107 the comparing prism p beneath the slit or out to one side. T attached to a right and left threaded spindle serves to widen or narrow the slit s. Attached to the upper part of K is the re- mainder of the eyepiece with its eye lens e vertically movable by rack and pinion through the milled head F. Fitting above e is FIG. 54. Microspectroscope. (W. & H. Seibert.) a tube A carrying an Amici prism R consisting of three prisms of crown glass (UD = 1.534) alternating with two prisms of flint glass (n D = 1.587). Since the total deviation of a ray of light entering a series of prisms is equivalent to the sum of the deviations which would be imparted to it by each unit in turn, it follows from the alternate io8 ELEMENTARY CHEMICAL MICROSCOPY FIG. 55. Microspectroscope. arrangement of the glass prisms, three low and two high, that the deviation of the system will be the difference between the devia- tions produced by the crown and flint prisms. The net result is that for rays of medium wave length (yellow-green) the path of the emerging rays lies substantially in the same line as that of the USEFUL MICROSCOPE ACCESSORIES 109 rays entering the system, hence it is usual to term such a prism system, a direct vision prism. The dispersive power of such a system is equivalent to that which would be produced by the prisms of flint glass alone. In the diagram, Fig. 55, the total dispersion indicated is therefore not theoretically correct. The measuring device of the Seibert microspectroscope fits above the tube A. It consists of a diaphragm with a very tiny triangular opening I mounted in the sliding plate B and illumi- nated by the mirror n; an image of this opening is projected by the lens / as a tiny bright white triangle upon the inclined surface of the prism R and is then reflected to the eye at i. The knob L serves to slide the lens / and thus focus the image of the triangular opening. The plate in which the diaphragm is mounted can be displaced vertically by means of a micrometer screw; the amount of displacement is indicated upon the scale S and by the gradua- tions upon the drum g; one complete rotation of the drum (100 divisions) is equivalent to one division of the scale S. To facilitate the illumination of the diaphragm opening I, the mirror n is attached to a rotating collar t. The position of a line in the spectrum is ascertained by bring- ing the triangle image to such a position that the line bisects the vertical angle. The scale and drum divisions are then read and recorded. The equivalent of this reading in wave lengths is obtained from the calibration of the instrument by the method given below. Should the object, whose absorption spectrum is to be studied, be so small that its image fails to completely fill the length of the slit, the slit must be shortened until the object completely fills it and there will be no light reaching the eye which does not first pass through the object. This is accomplished by pushing the comparing prism into place, thus cutting the spectrum in half. At the same tune the mirror m is turned aside so that no light enters O. Should the image of the object still fail to fill the length of the slit, the sliding diaphragm d is moved toward the center by turning the head D, until the slit length is reduced to the proper dimensions. In order to center the object, examine and focus it, it is neces- 110 ELEMENTARY CHEMICAL MICROSCOPY sary to remove the tube A carrying the prism. 1 The slit s is opened to its full width and the microscope focused in the usual manner, the eyepiece having first been itself focused by means of F and set at the proper calibration reference mark c. Before the instrument can yield scale readings convertible into wave lengths, it must be calibrated. This will necessitate placing upon its tubes certain reference or indicator marks. The instrument is removed from the microscope tube M, pointed toward the sky and the slit narrowed. The spectrum should appear as a long rectangular band of colored light crossed by many fine black lines at right angles (Fraunhofer's lines) to its length. Should these lines appear inclined, the tube A must be turned slightly until they are made normal to the spectrum length. Having thus carefully adjusted the prism to the proper position with reference to the slit, make the reference marks b upon A and upon r in order to fix this position. Now carefully focus the spectrum by means of F, using the narrowest slit possible until the Fraunhofer lines appear sharpest. This should be done on a bright sunny day. Scratch the mark c to indicate this position. Turn t and tip the mirror n so as to reflect light into the tube and move L until a bright sharp white triangle is seen when looking into the eyepiece. Carefully turn the cap carrying the measuring device until the apex of the bright tri- angle takes a position just a trifle above the center of the spec- trum band. This position is easily ascertained by pushing the comparing prism in place beneath the slit; half the spectrum will now disappear. The most convenient position for the bright spot of light is when the base of the triangle falls just below the dividing line. Make the marks indicated at a so as to fix this position. The instrument is now ready for calibration. It can be taken apart at any time and the parts replaced so as not to alter the values of the scale divisions. After calibration, if, at any future time, wave length measurements are required, the 1 In other forms of microspectroscopes, as, for example, those manufactured by Zeiss, Leitz and others, the Amici prism is so mounted as to swing upon a hinge above the eye lens. This greatly simplifies adjustments. Unfortunately all of these instruments have measuring devices too crude to be of value to the chemist. USEFUL MICROSCOPE ACCESSORIES III instrument is first set so that all the reference marks take the same positions as when the spectroscope was first adjusted. Measurements of line or band positions are made by bringing the bright white triangle to such a position that the line or the edge of the band bisects the acute angle of the triangle. The scale S and drum g are then read and recorded. S reads from o to 10, g in hundredths of S. For example, in the instrument illustrated: Fraunhofer c = 0.42, D = 1.41, G = 7.11, etc. In calibrating by means of the Fraunhofer lines direct sunlight should be thrown into the instrument by means of the microscope mirror. For bright lines, hold the instrument clamped securely in place on a suitable clamp stand and direct it toward a Bunsen burner flame into which the metallic salts are to be introduced. The following lines will be found convenient for the calibration: Line. Corresponding wave length in Angstrom units. Line. Corresponding wave length in Angstrom units. A 7600 F 4681 K 7682 Sr/3 4.6O7 a 7201 CS.. AC f S B 6870 Cs/s.... 4 CO 2 Li 6708 d 4.383 c 6*63 G 4.308 Na (D) . . liquid, the emergent rays are converging. In Fig. 107 the solid line arrows indicate the direction of the moving mirror, while the dotted line arrows that of the corresponding direction of movement of the disk of light in the image. These FIG. 107. Oil Globule and Air Bubble illuminated with Oblique Light. (Gage.) diagrams indicate the behavior of the light rays, but in the image in the microscope positions and directions are reversed; hence the phenomena observed are those described above. It thus appears that under oblique illumination the contour bands are heavier or darker on one side of the image of the object than on the other, the particular side which is darker depending upon the difference in the indices of object and mounting medium and the direction of the illuminating rays. Advantage is taken of these facts to determine by means of oblique light whether an object whose refractive index is sought has a higher or lower index than that of the test liquid in which it is immersed. Ob- lique light 1 is obtained by swinging the mirror to one side when no condenser is employed, or by sliding a piece of black paper or card just below the condenser or by holding a finger just below the condenser so as to cut off about one-half the lower aperture. In the chemical microscope slide a piece of stiff black paper between the condenser and the ring attached to its lower part. The preparation on the stage will then be illuminated by oblique 1 See Wright, Oblique Illumination in Petrographic Microscopic Work; Amer. J. Sci. (4) 36 (1913), 63. THE DETERMINATION OF REFRACTIVE INDEX 189 light. The phenomena resulting can best be understood by consulting Figs. 108 and 109, in which the indicated directions of the passage of light rays have been greatly exaggerated. The crystal H has a higher refractive index than the liquid surround- FIG. 108. Contour Bands in Half Shadow Illumination. FIG. 109. Contour Bands in Half Shadow Illumination. ing it; the rays passing through are therefore convergent, but only those at the left can enter the objective O; hence, the left side is bright and the right side dark. But in the case of the crystal L whose index is less than that of the liquid the emerg- ing rays diverge, yet here again only part of the rays can enter the objective 0; in this instance those on the right; thus the right side is bright: the left dark or in other words, the opposite of the phenomena observed with crystal H. Conducting our observations with the condenser only very slightly lowered and the paper diaphragm inserted from the left until the dark shadow extends approximately to the center of the field, the phenomena seen will be as indicated in Fig. 109. The crystal H of higher index than the liquid appears dark on the dark side of the field and bright on the light side of the field; but the crystal fragment L of lower index than the liquid appears bright on the dark side of the field and dark on the bright side of the field. This is as it should be from Fig. 108, since in the image formed in the microscope the directions are reversed. If we now lower the condenser a reversal of all the above phenomena takes place. It is therefore always wise to check the results recorded with condenser raised by lowering the con- 190 ELEMENTARY CHEMICAL MICROSCOPY denser; moreover the phenomena are much more distinct with lowered condenser. There is little chance for an error of judgment if the student will start with condenser raised and stopped down, and first slowly raise the objective, noting the direction of apparent move- ment of the contour bands or halo. Next test with oblique light and note the relative position of the dark contours with respect to the dark half of the field and finally lower the condenser and test again with oblique light. All three of the sets of observa- tions should be in accord. The student should also learn to use a finger below the condenser to obtain oblique illumination and thus save time. Since most of the liquids employed for the determination of refractive index by the immersion method have a greater dis- persive power than the solids, at the end point in the immersion method the images usually appear surrounded by colored fringes. The conditions which usually obtain are that when the liquid and solid have the same refractive index for yellow-green rays, the liquid will have a higher n for blue rays than the solid but the solid will have a higher n for red rays than the liquid. It follows that the emerging red rays will be convergent as diagramed in S, Fig. 108, while the emerging blue rays will be divergent. 1 No dark contour bands will be sufficiently prominent to be noticeable, but the image will exhibit a bluish fringe on the outside and a reddish fringe on the inside, or with oblique light bluish on one side, reddish on the other. Raising the objective will cause the red fringe to move inward and the blue fringe outward. It is evident that this color dispersion phenomenon enables us to still further assure ourselves when we have found the liquid having the same n as that of the solid under examination. When in the course of the experiments a marked color fringe is seen with the absence of black bands, the point has been reached in which liquid and solid have the same refractive index for light rays of medium wave length. To obtain more accurate results recourse must be had to monochromatic light. In preparing a series of liquids of regularly differing refractive 1 Wright, Amer. J. Sci., loc. cit. THE DETERMINATION OF REFRACTIVE INDEX 191 I indices for use in this immersion method, it is advantageous to select those having a slightly greater color dispersion than will be found in the solids to be tested. But highly dispersive liquids must be avoided since the color bands or halos are then so marked as to seriously interfere with the recognition of dark contours. Having ascertained as described above whether the crystal fragment has a higher or a lower index than that of the liquid first tried, and thus in which direction to proceed, a second liquid whose index is probably very much nearer that of the solid is chosen. The first liquid is carefully removed by absorbing it with a bit of filter paper, a drop of the liquid next to be applied is added and allowed to flow completely around the crystal; after standing a few moments this is removed as before and a new por- tion added. The preparation is tested by raising the objective and by the half-shadow method to learn whether the solid or the liquid has the higher index. The process is repeated until the proper liquid has been found. In making the trials add first a liquid of a higher then one of lower value. When sufficient solid material is available it will be found that time will be saved and much more reliable data obtained if an entirely new preparation is made with each liquid. This also avoids wasting valuable liquids. At the end of the chapter will be found tables l of liquids for use in the determination of refractive indices. In Table IV will be found the indices of isometric crystals useful in estimating the refractive indices of liquids. If it is found that the index of no liquid in a series at hand cor- responds to that of the crystal under observation, mixtures of two liquids may be made and the index of refraction of the mix- ture can roughly be estimated from the formula, 2 *hVi + %V 2 = n (V, + V,), in which Vi and 2 are the volumes of the two liquids, n\ and n^ 1 For exceptionally complete lists of media for refractive index determinations see Johannsen, Manual of Petrographic Methods. 2 Schroeder \an der Kolk, Mikroskopische Krystallbestimmung, Weisbaden, 1898, p. 13. IQ2 ELEMENTARY CHEMICAL MICROSCOPY their respective indices of refraction and n the refractive index of the mixture. It is obvious that in the preparation in this manner, of liquids of intermediate index values, it is essential that the two liquids shall be miscible in all proportions, and that no new chemical compound shall result from the mixing. Since a determination of refractive index may often require a period of time sufficiently long to result in an appreciable loss of liquid through evaporation, the liquids chosen for mixing should, theoretically, not differ greatly in their boiling points, otherwise there is a possibility of the concentration of the less volatile liquid increasing. It will also be obvious that in order to obtain suffi- ciently exact calculated values from the equation given above, the liquids mixed should not have widely different densities. For these reasons approximate boiling points and densities have been given in Table I. This formula for calculating the index of refrac- tion of a mixture of two liquids is based upon the assumption that to the final mixture each component contributes equally its own proportional part of the final index. There seems to be no sound theory in support of this assumption nor do the facts appear to be in accord with the formula. From experiments made in the Cornell laboratories, using an Abbe refractometer to determine the refractive index of the liquids before and after mixing, it was found that the calculated results were not always dependable to the second decimal place. The first decimal was always correct, the second usually so, but very rarely indeed would the third agree in calculated and found values. Mixtures allowed to stand several days and again measured gave similar results, showing that equilibrium had been reached when the first obser- vations had been made. Formulas of the kind given above should not be employed if the same degree of accuracy in calculation is wanted, as the im- mersion method will yield in practice. The immersion method above described permits an accuracy in the determination of the refractive index within 0.005 but with monochromatic light and more refined methods of illumi- nation an accuracy of 0.002 or even o.ooi db may sometimes be reached. THE DETERMINATION OF REFRACTIVE INDEX 193 The Refractive Index of Anisotropic Substances. Crys- tals are either isotropic or anisotropic. In isotropic crystals light rays are refracted to an equal degree, no matter in what direction through the crystals the rays are sent, since the velocity of transmission of light is the same in all directions through the crystals, providing the crystals have not been subjected to stresses or strains. In the determination of the refractive indices of isotropic crystals, it is obvious that the same value will be obtained in all directions through the crystals. In the case of anisotropic crystals, however, the rate of transmission of light is different in different directions through the crystals. In order to better appreciate the influence of these properties upon the refractive index, it is necessary to briefly consider a few funda- mental facts. A ray of light, when passing obliquely from one medium into another whose rate of transmission for light rays is different, will be deflected from its original path according to the equation C1T1 / \f - = . , in which i is the angle formed by the incident ray sin r V and the normal, r the angle formed by the refracted ray and the normal, and V and V the velocities of the transmission of the light in the two media. When the rays pass from a medium having a higher rate of transmission into one of lesser rate the deflection is toward the normal, but when passing from a medium with a lesser rate into one of higher rate the bending is away from the normal. In microscopic work the light rays are usually passing from air into a denser medium. If in the above equation we assign to the velocity of light in air the value of i, the equation ' sin i i , , sin i . ,, r , , . , r becomes - = . , but - is the expression for the index of sin r V sin r refraction, from which it appears that the refractive index is inversely proportional to the velocity of the transmission of light in the medium. Since in anisotropic crystals, the rate of trans- mission of light rays differs according to the direction through the crystal in which the rays are sent, it is obvious that the re- fractive index of an anisotropic crystal cannot be expressed by a single value and further, that of the several values given by a IQ4 ELEMENTARY CHEMICAL MICROSCOPY doubly refracting crystal, the greatest index will be found in the direction through the crystal of the lowest rate of light trans- mission and the smallest index in the direction of the highest rate of light transmission. In other words, different values for the index of refraction will be obtained according to the position in which the crystals lie upon the stage of the microscope. Crystals belonging to the tetragonal and hexagonal systems (uniaxial crystals) possess two indices. Crystals belonging to the orthorhombic, monoclinic, and triclinic systems (biaxial crystals) have three indices. In uniaxial crystals one value corresponds to that given by the ordinary ray and the other to that given by the extraordinary ray. The first value is found in that direction through the crys- tal where the light vibrations are transmitted transverse to the vertical crystallographic (and in this case optical) axis and is designated by the Greek letter o>; the second value is observed when light is transmitted through the crystal parallel to the verti- cal axis. This index is designated by the Greek letter e. The double refraction of uniaxial crystals is said to be strong when o> is greater than c, and weak when the reverse is found. When the refractive index co is greater than e, the crystal is said to be opti- cally negative and when less than e, optically positive. Some crystallographers prefer to designate the two refractive indices by the letters a and 7. In this case 7 a expresses the strength of double refraction and when a is greater than 7 the crystal is optically negative. 1 In biaxial crystals three different values for the rate of light transmission can be found, or in other words biaxial crystals have three axes of elasticity or directions of vibration; the axis of maximum rate of vibration transmission is commonly designated by the German letter a ; that of the minimum vibration by c and the intermediate axis by b. Since there are three axes of elas- ticity, three different values for the index of refraction may be obtained, the smallest value a in the direction of the maximum 1 In order to be sure of the values for co and e, a number of different crystals should be tried out. co will be constant in all of them, e will differ slightly accord- ing to the position of the crystals. THE DETERMINATION OF REFRACTIVE INDEX 195 axis a, the greatest value 7 in the direction of the axis c and an intermediate value in the direction of the b axis. The double refraction of the crystal will be strong or weak according to how much greater 7 is than a. To determine whether a biaxial crys- tal is optically positive or negative requires other data than refractive indices. In uniaxial crystals the determination of which index is o> and which e is comparatively simple since e coincides with the crystallographic c axis; but in the case of biaxial crystals it is seldom that a chemist possesses either the knowledge or a microscope sufficiently well equipped to definitely locate the different axes of elasticity, since their directions are indicated by neither the crystallographic nor the optical axes. For this reason it is wiser for the chemist-analyst to follow the methods of Kley, 1 Bolland 2 and others, and record values as obtained in the method given below. Swing the polarizer in place, having first removed all condens- ing lenses. Place upon the stage an object slide carrying the crystals or crystal fragments to be examined immersed in a liquid of known refractive index and covered with a tiny thin cover glass. Place the analyzer over the eyepiece (or slide it into the tube if an instrument of this type is used) and set the graduated circles of both prisms at zero so that their planes of vibration are crossed. Turn the stage of the microscope until the crystal selected for observation extinguishes; remove the analyzer. Ascertain by raising the objective whether the index of the crystal is greater or less than the liquid; check results by oblique light by placing the finger part way across the opening of the polarizer. Substitute one liquid after another until the refractive index of the crystal is ascertained, being very careful not to alter the position of the crystal. If the crystal is moved replace the analyzer and readjust the crystal to the position of extinction. Read the position of the crystal as indicated on the circumference of the stage and rotate the stage so as to turn the crystal exactly 90 degrees to its position of extinction and proceed 1 Kley, Zeit. anal. Chem., 43 (1904), 160. 2 Bolland, Monats., 29 (1908), 991; 31 (1910), 387. 196 ELEMENTARY CHEMICAL MICROSCOPY with the determination of the refractive index just as before. The two values obtained will, in the case of uniaxial crystals, be the indices e and co. When dealing with biaxial crystals in order to use the values in Bolland's tables first set the crystal so that its prism edge lies parallel to a plane passing through the short diagonal of the polarizing nicol. Next determine the index for a position at 90 degrees to the first. If a third value can be found, determine it If the values for a and 7 are wanted, determine the values for a very large number of fragments; the minimum value will be a and the maximum 7. Determination of the Refractive Index of a Liquid by the Method of the Displacement of Images. When an object is viewed through a liquid from a point in a line normal to the plane in which the object lies, the image observed will appear to lie in a plane above that of the object, the amount of displacement being dependent upon the refractive index of the interposed medium. 1 FIG. no. If, therefore, we place a liquid in a cell of depth DD' (Fig. no) and measure the amount of displacement of image OO' of a mark at O upon the lower surface of the glass slide, the index of re- DD' fraction n will be found from the equation n 1 This method is very old and is generally known as the Due de Chaulnes Method, having been described by him in 1767-1770. See also Sorby, Chem. N., 37 (1878), 151; Watson, Physics; Johannsen, Manual of Petrographic Methods. THE DETERMINATION OF REFRACTIVE INDEX 197 Method. Cement upon a thin object slide of clear glass, a ring, whose top and bottom are ground true and parallel. The cell thus formed should be approximately one millimeter deep and several millimeters in diameter. Place in the cell the liquid whose refractive index is sought and cover with a thin cover glass of greater diameter than that of the cell, as shown in Fig. no. Determine the thickness of the liquid layer by means of the graduations on the fine adjustment as follows: focus carefully upon the upper surface of the object slide. Read the position of the fine adjustment. Slide the cell along until the projecting cover glass is in the field and focus upon the under side of the cover glass and record this value. This will give the depth of the liquid plus an error due to the refraction of the cover glass. Next focus upon the upper surface of the cover glass. The dif- ference between the last two readings will give the apparent thickness a of the cover. The true thickness x of the cover x can now be found from the equation 1.52 = - where 1.52 is the Of refractive index of glass, and x a will be the value to be sub- tracted from the thickness of the liquid layer found above in order to obtain the true thickness. Call this corrected thickness A. Now push the preparation along so that only the object slide appears in the field. Focus sharply upon the upper surface. Without disturbing the adjustment push the slide to one side so that it is no longer in the field and by means of the Abbe con- denser project the image of a net ruled scale into the plane of the upper surface of the object slide. (See page 156.) This is done by adjusting the Abbe condenser until the scale is very sharp and clear in the microscope. Under no circumstances must the focus of the microscope be changed. Push the slide into the field so that an observation may be made outside the cell but through both slide and cover glass, that is in the line MI. The image of the net rulings will be faint and hazy. Read the fine adjustment and focus up by means of it until the rulings become as sharp and clear as in the first place. There is thus obtained the amount of the displacement of image due to slide and cover glass. Without changing the focus, move the preparation until 198 ELEMENTARY CHEMICAL MICROSCOPY an observation may be made through the filled cell, i.e., in the vertical line M 2 . The image of the net will now either be in- visible or badly out of focus; having recorded the reading on the fine adjustment, again focus up until the image of the net be- comes sharp and clear; read the fine adjustment. This value is the amount of displacement due to slide, cover glass and cell contents. The difference between the first reading obtained and the second gives the amount of displacement OO' of the image O due to the liquid in the cell; subtract this last value from the depth of the cell A; the remainder 5 equals O'D. The refractive index of the liquid is therefore n = o Providing great care is exercised in the micrometric measure- ments the determination of the displacement of image due to the object slide and cover glass may be eliminated as follows: Pro- ject the image of the grating into the focal plane with no slide in the field, move the slide until an observation can be made through both slide and cover glass (vertical line MI), set the micrometer of the fine adjustment at zero and focus the plane of the net by means of the screw adjustment of the substage condenser; the dis- placement of the image due to slide and cover glass has thus been eliminated. Without further changing the focus of the optical systems either above or below the stage, move the cell containing the liquid so that an observation can be made through the center of the cell (vertical plane M 2 ). Focus up with the fine adjust- ment; the reading of the scale ; will give the displacement O'O, /. 5 = A - O'O and n = - - o In all cases where measurements are made by means of the fine adjustment, first turn the graduated head until the pillar of the instrument is raised sufficiently to allow for a liberal move- ment up and down in focusing. A number of readings should always be taken of the position of the focal planes and the results averaged, never forgetting to lower the objective slightly below the position of the sharpest focus and then raise it until the image appears most sharply defined, thus avoiding the error due to " back-lash." THE DETERMINATION OF REFRACTIVE INDEX 199 It is obvious that the cell must be accurately ground in order that the cover glass shall lie parallel to the object slide, or if not truly parallel, that the measurement of the depth of the cell and that of the displacement of the image be made at the same point. Since there is always a thin film of liquid between the cover glass and top of the cell, the value for A should be determined with the cell filled and all data necessary for the computation be made at once. This method gives values to three decimals for n of which two places at least will be correct and the third not far from the true value. Correct results are more easily obtained with red or yellow light than by ordinary daylight. In the absence of a suitable cell, a- simple container for the liquid may be made from narrow strips of glass cut from an ordi- nary thin object slide and laid as shown in Fig. in. These strips of glass are easily cut with a glazier's diamond or with the sharp end of a file. The liquid to be studied is allowed to drop into the opening between the glass strips, and FlG iri the cell upon being covered remains filled by capillarity. The cover is gently pressed down and the excess of liquid removed with absorbent paper or a piece of drawn out glass tubing. Since there is a film of liquid in this case between both the upper and lower surfaces of the cell walls, considerable care must be exercised to avoid serious error. In any event the results are to be regarded as approximations only. A number of other more accurate methods for the microscopic determination of the refractive indices of liquids have been proposed, but these require specially constructed prisms, wedges or lenses, or fragments of glass of known index of refraction. For information as to methods, apparatus and accuracy the student is referred to the excellent paper by F. E. Wright, The Measurement of the Refractive Index of a Drop of Liquid, Journal Washington Academy Sciences 4, (1914), 269. 200 ELEMENTARY CHEMICAL MICROSCOPY Determining Thickness by Displacement of Image. It is obvious from the above discussion that if we have a transparent body whose refractive index we know, we can determine its thickness by applying similar methods. Supposing in the dia- gram, Fig. 1 10, we are dealing with a solid body. Its thickness will be T = n O'D. In this case the value of n is known, and O'D can quickly be ascertained experimentally. The value for T thus found will be accurate within approximately 0.02 mm. In the absence of a cover glass gauge, the thickness of cover glasses or of object slides may be thus determined: place a tiny, very thin drop of ink upon the upper and upon the lower sides of the glass plate, so that they fall almost in the same line; focus first upon the lower surface of the glass, using the ink spot as a guide, read the fine adjustment and focus up until the upper sur- face of the slide is in focus, again read the fine adjustment; the difference between the two readings gives the displacement of image. Taking for the value of n for cover glasses and ordinary object slides 1.52, the thickness is readily calculated from the formula given above. Glass varies according to its composition from n = 1.52 to n = 1.59. For quartz, n = 1.544 to 1.553. THE DETERMINATION OF. REFRACTIVE INDEX 2OI TABLE I. LIQUIDS FOR THE DETERMINATION OF THE REFRACTIVE INDICES OF SOLIDS BY IMMERSION METHOD. Index of refraction. 1 Name. Approximate boiling point. C. Approximate density. 32 .36 37 39 .40 44 46 .46 47 47 47 .48 49 49 50 Si 52 55 .56 :3 58 .61 .62 .62 .625 .63 65 . 7 6 2 9S 3 (?) Methyl alcohol 66 35 78 98 132 61 76 174 290 155 0.79 0.71 0.79 0-73 0.83 1.48 I-59* 0.92 1.61 0.86 0.91 0.96 0.86 0.88 0.98 .04 .20 49 .00 '83 05 83 .09 .29 So 50 3-34 1. 12 n=i. 3 6 2 n=i.49 s w = i.5o 2 w=i.S4 8 n = i.S9 3 n=i.64 3 w=i.8-h Ethyl ether Ethyl alcohol Heptane Amyl alcohol Chloroform Carbon tetrachloride Cajeput oil Glycerine Turpentine Olive oil Castor oil Xylene 136 80 Benzene Clove oil . ... Cedar wood oil Monochlorbenzene Nitrobenzene 132 209 155 197 195 149 240 187 239 46 255 277 1 80 272 Monobrombenzene Orthotoluidine Monobromphenol . . Bromoform Quinaldin Jvlonoiodobenzene Quinoline Carbon bisulphide Alpha-monochlornaphthalene Alpha-monobromnaphthalene . Methylene iodide Phenyl sulphide 1 The values for w in this column are those obtained in the author's laboratory at 20-22 C. by means of the refractometer on Merck products. 2 Schroeder van der Kolk, 1. c. Kley, 1. c. 202 ELEMENTARY CHEMICAL MICROSCOPY TABLE II. LIQUIDS FOR DETERMINATION OF REFRACTIVE INDICES OF MINERALS, CRYSTALS, ETC. WRIGHT'S SERIES. Bui. 158, Carnegie Institute. For indices Use mixtures of from to 1-450 1.480 1-475 1-535 Petroleum and turpentine. Turpentine and ethylene bromide or turpentine * and clove oil. 1-540 1-635 Clove oil and alpha-monobromnaphthalene. 1 .640 1-655 Alpha-monobromnaphthalene and alpha-mono- chlornaphthalene . 1.66 1.740 Alpha-monobromnaphthalene and methylene iodide. 1.740 1.790 Sulphur dissolved in methylene iodide. 1.790 i .960 Methylene iodide, antimony iodide, arsenic sul- phide, antimony sulphide, sulphur. This series requires the use of but few liquids and keeps the dispersion of the liquids within narrow limits throughout the series. As prepared for use, each one of the series should differ from the next above or below by 0.005. The value of n in each mixture made must first be de- termined by means of a refractometer. TABLE III. MEDIA FOR REFRACTIVE INDEX DETERMINATIONS. 1 Weighing out and grinding together in a mortar the weights of the substances given in the table, a series of eutectics is obtained, each of which will have the refractive index indicated in the first column. Checking with a refractometer is unnecessary. Refractive index. Components in grams. .487 SOS 535 54 55 56 57 58 59 .60 .605 Tl Tt Sa lymol. ... 35 lymol 67 lol 60 60 .60 Cam phor. . . 65 33 ... 40 ... 40 ... 40 ... 40 ... 40 ... 40 ... 40 ... 40 ... 40 Alpha-E amine aphthyl- 5 14 24 34 44 60 82 IOO 60 60 60 60 60 . 60 Merwin, J. Wash. Acad. Sci., 3 (1913), 35- THE DETERMINATION OF REFRACTIVE INDEX TABLE IV. 203 COMPOUNDS BELONGING TO THE ISOMETRIC SYSTEM WHOSE CRYSTALS MAY BE USED FOR THE DETERMINATION OF THE REFRACTIVE INDICES OF LIQUIDS. Refractive index. 1 Name. Formula. 439 450 459 481 485 .490 494 515 544 553 559 .566 571 .640 645 .650 .657 .667 .698 .700 755 1.788 I-95+ 2.071 Sodium alum Potassium alum Ammonium alum Potassium chromium alum . Ammonium iron alum Potassium chloride Rubidium chloride Sodium chlorate Sodium chloride Rubidium bromide Potassium bromide Strontium nitrate Barium nitrate , Ammonium chloride Cesium chloride Rubidium iodide Potassium chlorostannate . Potassium iodide Cesium bromide Ammonium iodide Arsenic trioxide Cesium iodide . . Lead nitrate. . . Silver chloride Na 2 SO 4 Ala (SO 4 ) 3 24 H 2 O K 2 SO 4 Al, (SO 4 ) 3 24 H 2 O (NH 4 ) 2 SO 4 A1 2 (SO 4 ) 3 24 H 2 O K 2 SO 4 - Cr 2 (SO 4 ) 3 24 H 2 O (NH 4 ) 2 S0 4 Fe 2 (S0 4 ) 3 24 H 2 O KC1 RbCl NaClO 3 NaCl RbBr KBr Sr (N0 3 ) 2 Ba (N0 3 ) 2 NH 4 C1 CsCl Rbl K 2 SnCl 6 KI ( n lies between \ 1.69 and 1.71 NHJ As 2 3 ^ T ( Bolland gives Csl .......... CsBr.. Pb (NO 3 ) 2 . . . . AgCl Groth gives i . 782 1 Most of these values for n are taken from Groth's tables. Leipzig, 1906-10. Chemische Krystallographie, 2O4 ELEMENTARY CHEMICAL MICROSCOPY TABLE V. REFRACTIVE INDICES AND CHARACTER OF DOUBLE REFRACTION OF TYPICAL CRYSTALS. Name. Formula. Crystal system Refractive index, 1 V Double refrac- tion. aorw. or . y Ammonium nickel sulphate Ammonium oxalate (NH 4 ) 2 Ni(SO 4 ) 2 -6H 2 O (NH 4 ) 2 Cj0 4 .H 2 (NH 4 ) 2 S 2 8 BaCl 2 2 H 2 O CuSO 4 5 H 2 O MgS0 4 7 H 2 O HgCl 2 Hg (CN), K (SbO)C 4 H 4 O,-H 2 O H 2 KAsO 4 K 2 Cr 2 7 K 2 Ni(SO 4 ) 2 -6H 2 O KNO S K 2 S 2 0, K 2 S0 4 AgNO, NajB 4 O 7 10 H 2 O NaNO, Na s PO 4 12 H 2 O HNatPO*i2H,O Na 2 S 2 3 -sH 2 Sr (SbO) 2 (C 4 H 4 0,) 2 C 12 H 21 O n H 2 C 4 H 4 0, CO (NH 2 ) 2 ZnSO 4 7 H 2 O M O M M Tr T T Tr M O Tr O M H H M M H M M T O .489 438 -498 .635 514 432 74 .65 .619 57 .72 .484 335 .461 493 .729 .446 58 44 432 .488 .638 -538 583 .485 .46 498 547 .502 .646 536 455 71 .60 .636 52 74 .492 505 .467 494 ^469 33 45 436 .508 .587 .566 .566 .61 48 .508 595 .587 .660 543 .461 .72 '637 .82 .505 .506 .566 498 .788 472 437 536 571 571 1.49 + + + + + + f + + + + + + Ammonium persulphate Barium chloride Copper sulphate Magnesium sulphate Mercuric chloride Mercuric cyanide Potassium antimonyl tartrate . . Potassium arsenate Potassium bichromate Potassium nickel sulphate Potassium nitrate Potassium persulphate Potassium sulphate Silver nitrate Sodium borate (tetra) Sodium nitrate Sodium phosphate (tertiary) Sodium phosphate (secondary) . Sodium thiosulphate Strontium antimonyl tartrate . . Sucrose Tartaric acid Urea Zinc sulphate 1 Values for n have been taken from Groth's tables and checked in the laboratory. For uni- axial crystals the first column is w and the second . For biaxial crystals the first column is , the second and the third y. REFERENCES. Tables of refractive indices in the following articles will be found by the analyst of great value in the identification of compounds by means of the immersion method. Schroeder van der Kolk Tabellen zur mikroskopischen Bestimmung der Mineralien nach ihren Brechnungsindex, Zeit. anal. Chem., 38 (1899), 615. Kley Ein Beitrag zur Analyse der Alkaloide, Zeit. anal. Chem., 43 (1904), 160. Bolland Die Brechnungs indices der weinsauren Alkaloide nach Einbettungs- methode. Monats., 29 (1908), 991. Die Brechnungsindices krystallinischer- chemischer Individuen nach der Einbettungsmethode von Standpunkte der ana- lytischen Praxis. Monats., 31 (1910), 387. CHAPTER X. QUANTITATIVE ANALYSIS BY MEANS OF THE MICROSCOPE. Some of the most difficult problems with which the chemist has to deal are those requiring an opinion as to the probable per- centage composition or amount of adulteration in materials which cannot be chemically analyzed. As typical examples of these cases may be cited, mixtures of starches, meals, adulterated flours, spices, teas and other food products; mixtures in which " firsts" have been sophisticated with an inferior quality of the same material; adulterated pigments; mixtures of wood pulps, paper pulps, textile fibers, etc. In the solution of problems of the above type there are several possible methods of procedure, but only two need occupy our attention. That these methods may be sufficiently accurate for our purpose the following requirements must be met. The components of the mixture must differ sufficiently in their appearance under the microscope to permit their easy recogni- tion, or they must be readily differentiated by their different behaviors towards stains or reagents; the components must not differ materially one from the other in specific gravity and must be small enough in size to allow mounting on an object slide and covering with an object glass; if of different specific gravities, their specific gravities must be known. We may (i) compare preparations made from the mixture of unknown percentage composition with preparations made from similar mixtures of known percentage composition which have been carefully prepared in the laboratory; or (2) we may by micrometric measurements ascertain the volume of the com- ponent whose percentage in the mixture is sought and from its known density compute its weight and hence its per cent in the mixture; or (3) in the case of solids made by fusion where the melt on freezing has been found to give rise to phases sufficiently 205 206 ELEMENTARY CHEMICAL MICROSCOPY characteristic in appearance yet differing according to the per- centage composition, the recognition of these crystalline phases will serve to indicate the probable composition of the mass. The last method (3) is restricted to materials such as alloys or related substances. An expert, knowing the characteristic appearance following certain treatments, is able, on studying materials of known components but of unknown percentage, to decide upon the probable proportion of the chief constituents without the necessity of a quantitative analysis. This type of analysis by means of the microscope can be practiced only by experts after long study and investigation and cannot therefore be here discussed. The first method may be employed in the quantitative analy- sis of all mixtures consisting of individual particles, fragments or crystals, which are not too large for microscopic examination, providing the component particles differ sufficiently in appear- ance to permit of identification and that mixtures of known per- centage composition can be prepared in the laboratory. Since this method has its chief application in estimating the amount of adulteration in a substance, the discussion will be confined to this aspect only. Method I. Prepare three standard mixtures containing the same components as the commercial products to be examined. In preparing these standards the adulterant must be carefully weighed out and added to a definite weight of the pure product; after thorough mixing, three mixtures of known per cent of adul- teration are thus obtained. From each one of these standards in turn, several like portions are taken, placed upon glass object slides in a drop or two of suitable medium (usually glycerine and water i : i), 1 distributed uniformly in the mounting medium and covered with a square cover glass, care being taken to avoid air bubbles; use just sufficient mounting medium to ensure an even distribution of the material throughout the whole area covered by the cover glass 1 Smith, Health Mag., 5 (1898), 286, has shown that in the case of starch mixtures a mounting medium of equal parts of glycerine, water and 50 per cent acetic acid is preferable. QUANTITATIVE ANALYSIS BY MEANS OF THE MICROSCOPE 207 and to completely fill the space below the confining cover yet not have a loss by the squeezing out of the liquid. One of the prepara- tions is then placed upon the stage of the microscope, and a count is made of the number of particles of the adulterant which are found in a field of the microscope. Having counted the foreign particles in several different fields, a second preparation from the same mixture is tried and so on until at least twenty or more q 5 123456789 10 11 12 13 14 Average number of foreign particles per field FIG. 112. counts have been made. A different mixture is then taken and the number of foreign particles determined exactly as in the first. Finally, the third known mixture is examined and counts made as before. Upon a sheet of "coordinate" paper lay out per cents of adulteration as ordinates and numbers of foreign par- ticles as abscissas. The averages of the counts of these particles obtained in each of the three mixtures of known per cent adul- teration are then marked upon the coordinate paper in their 208 ELEMENTARY CHEMICAL MICROSCOPY proper places, and a line is drawn through the zero and the three points; the "plot" obtained will be substantially a straight line if the work has been properly done. If the points laid out show a marked deviation from a straight line the components differ materially in their densities, or an error has been made. There is thus obtained a device, Fig. 112, by which we can determine, from a count of the foreign particles in any similar mixture, the per cent of this foreign matter present in material of unknown percentage composition. 1 FIG. 113. FIG. 114. Net Ruled Eyepiece Micrometers. FIG. 115. To facilitate counting an eyepiece with net micrometer is essential. Rulings are usually of two types, as shown in Figs. 113 and 114. Where type 113 is employed the entire field of view may be counted but in type 114 it is better to call a "field" that area comprised within the ruled square. This system is preferable to that of employing a cell with ruled bottom referred to below. An attachable mechanical stage will be found to be a great help in avoiding the making of counts in the same area more than once. Although the method just described appears at first sight to be crude and unreliable it has been found after a number of years' trial in the hands of a large number of students to yield excellent results. In the case of starch mixtures, where the foreign component 1 Chamot, Seventh International Congress Applied Chemistry, Section VIIIc (1909), 249. QUANTITATIVE ANALYSIS BY MEANS OF THE MICROSCOPE 2OQ is present in the proportion of 3 to 7 per cent the results found are very close to the actual per cent, but when 7 per cent is reached, the beginner has trouble in obtaining reliable counts, and above 10 per cent the method requires great manipulative skill. It must, however, be borne in mind that a method of this sort even at its best gives merely a close approximation to the true value. The chief difficulties which will be encountered are those of removing equal amounts in every case upon the end of a tiny spatula; of obtaining a uniform distribution of the material throughout the drop; and of lowering the cover glass upon the preparations without destroying the uniformity of distribution of particles or introducing air bubbles. A little practice, how- ever, will enable the analyst to work rapidly and accurately. If more nearly accurate sampling is desirable, a portion of the material is carefully weighed out, spread on a piece of glass or glazed paper in a thin square of as nearly uniform thickness as possible and then sampled by "quartering" in the usual manner 1 until a section equivalent to 2 to 4 milligrams is obtained for transfer to the object slide. An even better method consists in carefully weighing out a small portion of the material to be examined and mixing it with a known weight, several times greater, of a finely and uniformly powdered substance very soluble in water (or other solvent). After thorough mixing, a small portion of the preparation is removed, accurately weighed and transferred to an object slide. The selected mounting liquid is added, causing the soluble diluting solid to dissolve and disappear, leaving a known weight of the insoluble material under investigation evenly spread upon the slide. The number of foreign particles in this tiny portion can then be counted. In the case of most food products, such as starches, flour, meals, spices, etc., powdered sucrose, dextrose, lactose or soluble dextrine are most useful as diluents. When the mixtures under examination are of a density only very slightly greater than water and are insoluble therein, and 1 Kraetner, J. Am. Chem. Soc., 21 (1899), 659. 210 ELEMENTARY CHEMICAL MICROSCOPY therefore if suspended would subside only after a long period, it is possible to weigh out a portion of the mixture, add it to water, or better, water and glycerine, in a small graduated flask, fill to the mark, shake well and quickly remove one cubic centimeter or less, for counting. This method precludes an error arising from non-uniform, quantities but is longer and more cumbersome than the methods already described. FIG. 116. Object Slide Ruled in One-half Millimeter Squares. Instead of using a net ruled micrometer eyepiece some micros- copists employ a slide ruled in squares or a tiny cell with ruled bottom, as shown in Figs. 116 and ny. 1 The advantage of such devices of permitting the use of any eyepiece is usually outweighed by a number of undesirable features, chief among which may be mentioned the objections that the rulings on the slides are FIG. 117. Girard Counting Cell for the Analysis of Flour. not always clear when the particles to be counted are in focus; the relatively large size of the ruled squares with a high power; and the difficulty of properly cleaning the slides without event- ually injuring the rulings. When it is desirable to cover a definite area on the object slide it is far better to employ a micrometer disk-diaphragm 1 Made by Nachet et Fils, Paris, France. QUANTITATIVE ANALYSIS BY MEANS OF THE MICROSCOPE 211 properly calibrated and inserted into the eyepiece or to cut a square opening in a disk of dull black paper, thin card, metal or blackened mica, and drop the disk into the proper eyepiece by removing the eye-lens and allowing the disk to rest upon the diaphragm of the eyepiece. The proper size of opening is ascer- tained by eyepiece and stage micrometers, and a square hole of this calculated size is cut in the paper and the perforated disk is inserted in the eyepiece. The final adjustment is then made with the draw- tube. When the particles of material are of a sufficiently low density to remain suspended for a few seconds and one cubic centimeter portions can be removed the Sedgwick-Rafter counting cell used FIG. 118. Counting Cell. (After Whipple.) ' in the quantitative determination of the microscopic organisms in water may be profitably employed. This cell, Fig. 118, con- sists of a glass object slide of standard size to which is cemented a brass cell 5 centimeters long by 2 centimeters wide; its area is therefore 1000 square millimeters and being made exactly i millimeter deep, its capacity when closed with a cover glass is i cubic centimeter. Counts of particles are made in as large a number of fields as possible, using a net eyepiece micrometer 1 From The Microscopy of Drinking Water, by G. C. Whipple, p. 35, Third Ed. John Wiley and Sons, Inc. Reproduced here through the courtesy of the author. 212 ELEMENTARY CHEMICAL MICROSCOPY or an eyepiece with a central diaphragm opening adjusted to any convenient area on the slide. Results may be expressed either in numbers per cubic centimeter or in per cent by the plotting method described above. 1 In the biological examination of water the microscopic organ- isms are concentrated into a few cubic centimeters of water by a small sand filter contained in the stem of a funnel of special design. The sand, together with the supernatant small volume of water, is emptied into a test tube, given a rotary motion and as soon as the heavy sand subsides, the water containing the organisms in suspension is poured off and one cubic centimeter transferred to the counting cell. 2 Although used primarily for the purpose stated, this counting cell and method can be applied to many problems involving chemical analyses. In order to facilitate the counting and recording of the sus- pended matter found in water, Whipple has devised an eye- piece micrometer with special ruling. This type of micrometer has been found desirable as an aid in recording the size and number of masses of amorphous matter in water. By common consent American analysts have agreed to express these values in terms of the areas covered by the masses found in the cell. The unit employed is a square, 20 microns on a side, and there- fore equal to 400 square microns; this is known as a " standard unit." The eyepiece micrometer is ruled and so adjusted that with a given objective and eyepiece the smallest squares are equal to a standard unit, Fig. 115. Method II. When isolated particles of sufficiently definite shape can be found and are of known composition and density, it is possible to calculate their weight from micrometric measure- ments. 1 For further applications of Method I, see Meyer, Zeit. Nahr. u. Genuss, 17 (1909) 497: Ezendam, Zeit. Nahr. u. Genuss., 18 (1909) 462. Analysis of Starch Mixtures. Young, Bull, no, Bureau Chem., U. S. Dept. Agric.; Pollen in Honey. Boedemann, Landw. Vers. Sta., 75, 134; Smut Spores in Flour, etc. Oerum, Biochem. Zeit., 35 (1912), 18; Fat in Milk: Vauflart, Ann. Falsif., 4 (1911) 381; Analysis of Meals. 2 For details and precautions in water examination, the student should consult Whipple, The Microscopy of Drinking Water. New York, Wiley & Sons, Inc. QUANTITATIVE ANALYSIS BY MEANS OF THE MICROSCOPE 213 This method is especially useful in estimating the weight of substances imbedded in other materials in such a way as to be not easily separated; in the determination of poisons in forensic investigations; and in determining the weight of tiny metallic beads or pieces of metal, which, for one reason or another, can- not be weighed on a balance. The dimensions of the particles are first determined by any one of the micrometric methods described in Chapter VII. From these measurements the volumes of the particles are calculated and their weight then obtained by multiplying by the specific gravity of the substance. If the substance whose weight is to be determined can be made to take the form of a sphere the data found are usually as accurate as those obtained by weighing, but it is obvious that if only more or less irregular particles or crystals are available the method should be regarded as giving merely approximate results. Even so, the method must be recognized as of value since in many instances no other system of solving the problem of percentage composition may be available. This method of quantitative analysis by means of the micro- scope is very old and has been successfully applied to the deter- mination of gold and silver in fire assays (especially with the blowpipe) where the metallic beads obtained on cupellation are too small to weigh even upon a sensitive assay balance. With carefully fused beads it has been shown 1 that the results are accurate and quickly obtained. The first essential is that the little metallic globule shall be a perfect sphere. If it is not, it is placed in a tiny cavity in a piece of charcoal and fused before the blowpipe; after cooling, it is transferred to a drop of glycerine and water (i : i) on a glass object slide by picking it up with a drawn-out glass rod slightly moistened. Bring the metallic sphere under the center of the micrometer eyepiece, use an objective of low power, illuminate with axial light, with the Abbe condenser well lowered using a small diaphragm opening. Focus up slowly and as soon as the image reaches its maximum diameter record the scale 1 Goldschmidt, Zeit. anal. Chem., 16 (1877), 434- 214 ELEMENTARY CHEMICAL MICROSCOPY reading. Make several observations of the diameter of the sphere. Then illuminate the sphere by oblique light by swinging the mirror far to one side; determine the diameter again, making not less than three observations; the results should be the same as the measurements made with axial light. Average the results. The weight of the bead may now be calculated from the equa- tion W = (d z X 0.5236) s where d is the diameter of the sphere and s the specific gravity of the metal. 1 For the quantitative determination of minute particles of mer- cury micrometric measurements of the diameters of the globules of the metal and calculations of weight therefrom are also un- questionably one of the oldest and best methods at our disposal in toxicological examinations, in the analysis of mineral waters, urine, gases carrying mercury vapors, etc. Raaschou 2 has recently worked out in great detail the methods -and conditions essential for the quantitative separation of minute amounts of mercury from liquids. For details, the student should consult the original article. 3 When dealing ; with sublimates of metallic mercury consisting of so great a number of tiny globules as to render measurements of the diameters of all the globules impracticable, cause them to unite into a few large spheres by stirring the film with a fine needle, or stiff hair, or glass rod drawn down to a hair, but if this is done the needle or hair must always be examined with the microscope to see that no mercury has been removed by clinging to the stirrer. In order that accurate measurements may be made it is essential that the globules of metallic mercury shall never be so large that they become flattened and thus not perfect spheres. In determining the diameter of the spheres proceed exactly as described above, always making several measurements of the sphere diameters. From the average of the data thus obtained, calculate the weight W = (d 3 X 0.5236) X 13.59- 1 For gold, 5 = 19.33; silver = 10.4; platinum = 21.15; l ea d = 11-36; mer- cury = 13.59. 2 Raaschou, Zeit. anal. Chem., 49 (1910), 172. 3 See also page 319, Microchemical Detection of Mercury. QUANTITATIVE ANALYSIS BY MEANS OF THE MICROSCOPE 215 In estimating the percentage of the different fibers entering into the composition of a given sample of paper, it is customary in most commercial paper-testing laboratories to guess at the per cent of a given fiber without comparison with standards and without counting the fibers, the usual practice being for several analysts to "guess" at the composition independently. These men in time become very expert and their findings will generally check within one per cent. In the opinion of the author, com- paring with known standards, using the comparison microscope or comparison eyepiece is quicker and gives a more reliable approximation. Herzog 1 has suggested a microscopic method for the quan- titative estimation of the different fibers in fabrics, or for the per cent of different colored fibers in a fabric. Stated briefly, the process is as follows: A tiny piece of the fabric is imbedded in paraffin (M.P. 60) by repeated dipping. After cooling, sections about o.i to 0.2 millimeter are cut by means of a razor or microtome knife. One of the sections is transferred to an object slide, warmed until the paraffin melts and is tipped back and forth to evenly distribute the fiber fragments. A drop of balsam is placed upon a cover glass and lowered upon the preparation. The entire number of each different fiber is then ascertained by counting, using a net eyepiece micrometer. Having thus found the relative proportion of the fibers, their absolute size is next determined by measurements of length and thickness, or since the thickness of the section cut is known and the average diameter of different fibers is also well established, actual dimensional measurements may not be re- quired. The weight is calculated by multiplying the absolute size by the number of fragments and the specific gravity of the fiber. Quantitative microchemical methods with reference to the handling of minute amounts of material and weighing on a Nernst micro balance; the titration of tiny volumes of liquid; the measurement of tiny volumes of gas, etc., which do not re- quire the application of the microscope need no discussion here, 1 Herzog, Z. Chem. Ind. Kol., I (1907), 202. Z. Text. Ind., 1906, No. 4. 2l6 ELEMENTARY CHEMICAL MICROSCOPY since we are dealing solely with the application of the micro- scope to the solution of chemical problems. 1 Volume and Weight Per Cents from Area Measurements. The quantitative analysis of heterogeneous material in thin sec- tions through the determination of the areas occupied by the different components, as ascertained from their images when seen in the microscope, has long been employed by petrologists. The process is briefly as follows: The outlines of the areas of the component under consideration, in a given field of the micro- scope, are traced upon coordinate paper by means of a drawing camera; the value of a square of the paper is ascertained with a stage micrometer as hereinbefore described. The areas of the tracing may then be computed or may be accurately determined by means of a planimeter. Or the preparation may be photo- graphed with a coordinate (net-ruled) ocular in place, the value of the rulings in the image ascertained in the usual manner and the areas of the different component-sections in the photograph computed. 2 From the computed areas, volume per cents may be calcu- lated, and knowing the specific gravities of the components, weight per cents are easily ascertained. This method of quantitative microscopic analysis has recently been applied by Johnson to the examination of concretes. He has shown 3 that it is a simple matter to ascertain, whether, in a given concrete structure, a contractor has complied with the specifications as to proportions of sand, gravel and cement and further whether the material was properly mixed and wetted. Estimation of Molecular Weights by Micrometric Measure- ments. Barger 4 has described a most ingenious micrometric method whereby the molecular weight of a substance may be determined, providing a large enough amount of the material for weighing upon an analytical balance is available. A solution is made of known weight content of the substance 1 See Donau, Die Arbeitsmethoden der Mikrochemie, Stuttgart, 1913. 2 For further details as to rock analysis and for bibliography see Johannsen, Petrographic Methods, p. 290. 3 Eng. Record, Mar. 1915. 4 Barger. J. Chem. Soc. (London), 85 (1904), 286. QUANTITATIVE ANALYSIS BY MEANS OF THE MICROSCOPE 217 whose molecular weight is sought. A second solution of known strength is also made of a substance of known molecular weight. Drops of these two solutions are introduced alternately into a thin-walled capillary tube having a bore whose diameter is from i to 2 milHmeters. The tube should be 6 to 8 centimeters long. Between the drops which occupy a space about i to 3 milHmeters long there must be air spaces equal to approximately twice the lengths of the drops. The first and last drops should be those of the standard and from two to three times the length of the intermediate drops. After the drops are in place the capillary tube is sealed at both ends. The tube is then laid upon an object slide and cemented in place with Canada balsam or other suitable medium, the slide is then immersed in water in a suitable shallow vessel and placed under the microscope. By means of a micrometer the lengths of the drops are deter- mined and recorded in scale divisions but not in absolute units. After standing for about an hour measurements are again made. Owing to differences in vapor pressure, some drops have increased in length; others have decreased. The theory of the method is thus described by its author: "Each drop is placed between two others of a different solution, and can evaporate on either side into a small air-chamber. This chamber is soon saturated with vapor, which can condense freely on the drops. If the vapor pressures of the two solutions are equal the evaporation will equal the condensation, and there will be no change in volume of the drops. If, on the other, hand, the vapor pressures are unequal, there will be a gradient of vapor pressure in the air spaces; some drops will therefore be in contact with an atmosphere, the vapor pressure of which is greater than their own. Condensation will take place on these drops and they will increase. The others, alternating with them, will have a vapor pressure greater than that of the ad- joining air spaces; these drops will evaporate and thus decrease. Hence, there is a distillation from the drops of one series to those of the other series. By measurement we can tell which drops increase and hence ascertain which solution has the smaller vapor pressure. If the solvent is identical in both cases 2l8 ELEMENTARY CHEMICAL MICROSCOPY and if the solutes are non-volatile, the solution with the smaller vapor pressure will have the greater concentration of molecules and vice versa. " A series of tubes must be made in which the strength of the standard solution has been systematically varied in small frac- tions of a gram-molecule per liter. A tube is thus obtained in the series where there is little variation in the lengths of the drops of known and unknown or where there is change in the character of the variation, say from an increase in length to a decrease in length. It is evident that the molecular concentra- tion of the unknown must correspond to that of the known solu- tion at this point. Weight of unknown in grams per liter Concentration in gram-molecules found This may be made clear by quoting one experiment: Standard used, cane sugar. Unknown, glucose. Solvent, water. Molecular weight = Concentration of Standard in gram-molecules. Nature of change in length of drop of unknown. Tube i 0.05 Increase Tube 2 O. IO Tube 3 O. 12 Tube 4 O IT. Tube 5 o. 14 Decrease Tube 6 O. I 1 ? Tube 7 Tube 8 O. 2O 0.25 It is evident therefore that the concentration of the unknown material lies between the concentrations of tubes number 4 and 5, that is between 0.13 and 0.14 gram-molecule per liter. Hence, 192. That is, the i i i \ 25.02 25.02 molecular weight = -* 1 = 179, or -^ 0.14 0.13 molecular weight of the unknown lies between 179 and 192; the average = 185.5. Calculated for glucose, C 6 Hi 2 O 6 = 180. It appears from a very large number of experiments that this method is a simple and dependable one, apparently subject to errors no greater than those usually inherent in macroscopic molecular weight determinations. QUANTITATIVE ANALYSIS BY MEANS OF THE MICROSCOPE 219 As small amounts as 25 to 50 milligrams may successfully be used. For special precautions, sources of error and suggestions as to the choice of solvents and standards, the student is referred to the original article. This method of Barger's for the determination of molecular weights is another example of the manifold applications of the microscope. The microscopist whose laboratory is seldom equipped with apparatus for the determination of molecular weights by the usual methods of boiling or freezing points, or by vapor densities, may nevertheless obtain sufficiently accurate results for all practical purposes by the procedure outlined above. The method is worthy of far more attention by analysts than it has been given. CHAPTER XI. THE DETERMINATION OF THE MELTING AND SUBLIMING POINTS OF MINUTE PARTICLES OF MATERIAL. The determination of the melting point of a compound is usually one of the simplest and most reliable tests at our dis- posal for ascertaining the purity of a known compound or for obtaining an idea as to the probable nature of a substance of unknown composition. In the case of organic compounds the melting point is one of the first constants to be ascertained and even with certain inorganic substances a melting point deter- mination may often prove of great value. It not infrequently happens that such a small quantity of material is available that the usual laboratory methods are im- practicable and recourse must be had to some microscopic method of procedure. Often, the chemist deals with material containing a large proportion of amorphous matter mixed with a crystalline substance and a satisfactory separation cannot be effected; or again, a preparation is obtained in which there appears to be two or more different crystalline substances but no means for separating them can be found. In all these cases a melting point would give the needed information were it possible to effect a separation. By spreading out the material in a thin layer upon an object slide and examining the preparation with the microscope, we can almost always find crystals or fragments of material here and there not in direct contact with others, but appearing in the image isolated and free. We have thus in reality affected a separation and if we apply heat, we should be able to make reliable observations upon the behavior of each isolated particle. If in addition we have some means of controlling and measuring the heat applied, it is obvious that a melting point can be ascer- tained. Inasmuch as a variety of methods for temperature DETERMINATION OF MELTING POINTS 221 measurements are available, it follows that melting-point deter- minations may be obtained of material actually invisible to the naked eye. Furthermore, these determinations will, in most cases, be as accurate as those made by the usual capillary-tube sulphuric-acid method. Method A. (Approximate.) -* Where a series of pure com- pounds, readily crystallizable and each of known melting point is at hand the melting point of an unknown substance may be ascertained approximately by placing similar sized fragments of the known and the unknown side by side at the corner of a thin object slide. The rotating stage of the microscope is re- moved and a piece of asbestos board, perforated at the center, substituted as a stage. A bent glass or quartz tube drawn out to a jet at one end serves as a tiny burner and may be fastened temporarily to the substage ring. The tiny burner is so adjusted that the flame falls nearly in the line of the optic axis of the microscope. The slide carrying the material to be tested is placed under the microscope and focused and the tiny flame is very slowly brought nearer the preparation by means of the screw which serves to raise or lower the substage. The behavior of the material is watched very closely through the microscope, to determine whether the known or the unknown substance melts first. Other compounds of known melting point are tried until a known compound is found with which the unknown simultaneously melts or the unknown is found to melt between the melting points of two knowns. This indirect method is quick and convenient where mere approximations are needed. The operator after one or two trials soon learns to judge the temperatures given by the tiny burner according to the size of its flame and the distance below the slide. When comparing melting points in this manner first try the pure material with which the unknown is believed to be identical. Place the two substances so close together on the slide that when they melt, the molten masses will flow together; if they melt simultaneously and mix to form a homogeneous melt, the presumption is strong that the two fragments are of the same composition. If so, when the melt solidifies (freezes) a single component will result. 222 ELEMENTARY CHEMICAL MICROSCOPY Lehmann 1 long ago pointed out that this method of " fusion testing" could be made use of in qualitative analysis but the interpretation of the phenomena which may be observed, usu- ally requires a profound knowledge of chemistry and much practice in manipulation. In the Appendix will be found a table giving the melting points of compounds which can be employed in making estimations of melting points by the process described above. Method B. (Exact.) Melting points below the boiling point of water may be determined with great accuracy by means of a hot stage through which hot water is made to circulate. A convenient form of apparatus is shown in Fig. ng. 2 It consists FIG. 119. Apparatus for the Determination of Low Melting Points. of a glass box or trough, such as is commonly employed for the spectroscopic examination of liquids, the open end of which is provided with a wedge-shaped piece of rubber, forming a tight stopper. The hot water enters the cell through the glass tube A and escapes at B, the rate of flow being controlled by a stop- cock or screw-clamp. The hot water may conveniently be ob- tained by siphoning it through a small coil of copper pipe D heated by a Bunsen burner E. Or the heating system devised for providing a continuous flow of hot water through a Zeiss 1 O. Lehmann, Die Krystallanalyse, Leipzig, 1891. 2 Chamot and Albrech, Unpublished paper presented to the Cornell Section, Am. Chem. Soc.; May, 1906. DETERMINATION OF MELTING POINTS 223 butyrorefractometer may be employed. By regulating the heating flame and the rate of flow of hot water, very gradual or very rapid rises of temperature may be obtained or the temper- ature may be maintained almost constant. Jacketing the cell with asbestos simplifies the regulation of temperature. Heaters functioning on the principle of the thermo-siphon, Fig. 120, may also be employed for temperatures up to 85 to 90 C.; but above 90 de- grees the regulation of the height of the heating flame becomes rather diffi- cult and the sudden formation of steam usually results in a blow-off through the safety tube, in which the thermometer is only very closely in- serted. Substituting brine or oil for water, the temperatures can be raised to 125-150 degrees if the heating coil be used, but the author has never found hot oil to give satisfactory results in any thermo-siphon system, since the viscosity of the oil in the glass cell is too great to permit an even and sufficiently rapid rate of flow unless large conducting pipes be employed, necessitating a cell far too thick for use. The temperatures are most conveniently measured by means of a set of Anschiitz thermometers. Thermometers of this type are sufficiently small, so as not to project too far, and their graduations are such as to permit readings to be taken to o.i degree. A convenient arrangement for reading the thermometer and observing the melting point of the substance under observation is given below. With hot stages of the sort just described it is always a wise precaution to place the cell in a glass tray or shallow crystallizing dish to guard .against damage to the microscope should the hot stage break. FIG. 1 20. Heater for Melting Point Apparatus. 224 ELEMENTARY CHEMICAL MICROSCOPY Any flat surfaced, stoppered container may serve as a hot stage, as, for example, a small flat bottle. For temperatures above 150 C. the only convenient and uni- versally applicable heating system is by means of an electric current, resistance wire and suitable rheostat. The heating coil in this case may consist of manganin, nichrome or platinum wire. To obtain the best and most reliable results part of the heating coil should be above the object being heated and part below. Fig. 121 shows an electrically heated hot stage which has been in use in the author's laboratory for several years. It FIG. 121. Apparatus for the Determination of Melting and Subliming Points. consists of a low cylinder of "Alberene stone" closed at the top and bottom by thin glass, or by mica when high temperatures are employed. The heating coil H, H consists of fine platinum wire wound in fine coils. In the illustration A shows the Alberene stone; B, brass guides for the object slide acting as cover; C, adjustable wire fingers for supporting cover glasses, tiny crucibles, " micro" retorts, etc.; D is a removable, thin brass diaphragm cutting down the opening of the stage and serving as a radiator; T, thermometer; PP, binding posts; M, mica or glass window closing the bottom of the hot stage; and S, the object slide cover. The method of inserting the hot stage for use in place of the rotat- ing stage is shown in Fig. 122. By attaching an Abbe camera lucida to the microscope tube and properly tipping the mirror, the image of the scale of the thermometer may be so reflected as to be seen alongside of the material whose melting point is DETERMINATION OF MELTING POINTS 225 FIG. 122. Polarizing Microscope arranged for Observing Melting Points. 226 ELEMENTARY CHEMICAL MICROSCOPY to be determined. A lens attached to the body-tube or held in a separate stand serves to magnify the thermometer scale. It is thus possible to look into the tube of the instrument and to watch both the material and the thermometer. This arrange- ment and its applications will be readily understood by refer- ence to the illustration. With platinum wire coils a temperature somewhat higher than 700 C. may be obtained in the apparatus. The material to be tested may be either crystallized upon or supported on a small thin cover glass held by the wire fingers C or may be placed in a short piece, 5 millimeters long, of tiny thin- walled capillary tube fastened to the thermometer by a wire band. For ordinary materials these tubes are best held horizon- tally but for fats, waxes, etc., better results are obtained by slightly inclining the capillary and taking as the melting point the thermometer reading at the instant the fat slides out of focus. The melting point of anisotropic substances is sharply obtained by making the observations with crossed nicols and a selenite plate; the change from solid to liquid of tiny particles is thus remarkably clear since they vanish instantly on melting. The hot stage should in such cases be provided with glass windows. The upper window of the stage consists of a thin glass object slip (or one of mica or of quartz) held in place by the guides B, B, permitting sliding the cover. This is essential when dealing with materials which sublime, for in these cases the upper window becomes fogged with condensed material, and in such an event the cover is simply pushed along until a clear section is obtained. In determining melting points with any type of hot stage, it is obvious that the usual procedure should be followed, namely: make a preliminary observation and then start anew, raising the temperature very gradually as the melting point first observed is approached. Determinations of the subliming points of tiny particles may also be made by means of the hot stage. Electrically heated stages of several forms and for different DETERMINATION OF MELTING POINTS 227 ranges of temperature may now be had from several different optical firms. 1 The Determination of Subliming Points may be made in the hot stage illustrated in Figs. 121 and 122, or by the crucible method of Blyth described on page 243. 1 E. Leitz, Wetzlar, Germany, manufactures some especially convenient hot stages. For other types of hot stage see Cram, J. Am. Chem. Soc., 34 (1912), 954; and Cottrell, J. Am. Chem. Soc., 34 (1912), 1328. CHAPTER XII. METHODS FOR THE HANDLING OF SMALL AMOUNTS OF MATERIAL. Microchemical Methods. By microchemical methods we generally mean the application of chemical operations to the examination and study of very small quantities of material. The chief chemical operations with which we have to deal are: i. Solution; 2. Decantation; 3. Filtration; 4. Sublimation; 5. Distillation; 6. Precipitation; 7. Ignition, Fusion, and Mis- cellaneous Treatments. Since success in chemical microscopy requires skill in the technique of these operations each one will be discussed at length. i. Solution. Testing for Solubility. At the corner of a perfectly clean object slide of glass, quartz, or celluloid, place a small drop of water (or other solvent) ; the drop should be 3 to 4 millimeters in diameter and about i millimeter deep. Place close to this drop a tiny fragment of the material whose solu- bility is to be tested. Transfer the glass slip to the stage of the microscope and focus with a low power objective upon the edge of the drop nearest the fragment. See that the illumination, using an Abbe condenser, is carefully adjusted, and that the iris diaphragm is at least two-thirds closed. By means of a glass rod drawn out fine, a platinum wire or a stiff hair, slowly push the fragment into the drop, at the same time looking into the instrument so as to be able to note the phenomena which may take place the instant the material enters the solvent; for ex- ample, the substance may merely "melt" away, or it may de- crepitate, or give off bubbles of gas, or it may dissolve with decomposition (hydrolise), etc. A little practice is often neces- sary to enable the beginner to push substances into drops of solvent while looking into the instrument. It is of course 228 METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 229 necessary to remember that directions are reversed in the image formed by the microscope and seen by the worker, but if this is borne in mind there will soon be no difficulty in moving and turning objects while observing them through the microscope. If, after a few minutes, there appears to be no change in the appearance or size of the material being tested, warm the drop gently by holding it a second or two about one centimeter above the " reserve" or "pilot" flame of the laboratory burner (see Fig. 73, page 127). This tiny flame should be so regulated by means of the set screw as not to be over 5 millimeters high. Cool the preparation quickly by holding the slip for an instant in con- tact with a smooth metal block placed for this purpose near the burner, or, in the absence of such a cooling device, place the slide on the base of the microscope. Examine the fragment of material to be tested and note any change in its appearance and size. To heat a solution to boiling have a large drop at the very corner of a glass slip, tip the slip slightly so that the drop flows toward the corner and hold it so that the tip of the micro- flame (pilot flame) touches the glass just below the upper edge of the inclined drop. Watch closely and as soon as bubbles rise, remove from the flame and cool instantly by bringing in contact with a cool metal surface. It is necessary to work quickly, otherwise the evaporation will be so great that the preparation will become dry. Never place a hot slide on the stage of the mi- croscope, for the stage may be seriously damaged and the vapors arising will condense upon the objectives injuring them. Since the drop has been placed at the corner of the slide there is no danger of the glass cracking or breaking on heating, an accident that will almost invariably happen if the glass slip is heated at any other point than a corner. If quartz or platinum slips are used, heating at the corner is not essential to prevent breakage, but is more convenient. To determine whether any material has passed into solution, decant the liquid from the undissolved material (see Decanting below), and evaporate to dryness very carefully. In evapora- ting drops to dryness, never keep the material over the flame until all the liquid has been driven off. Simply warm the prepa- 230 ELEMENTARY CHEMICAL MICROSCOPY ration, then remove it from the flame and blow gently upon the warm drop, heat again and again blow; repeat the process until the solvent has been driven off. If this method is followed, a uniform, closely adhering film will result instead of irregularly distributed loose particles, and the danger of loss through de- crepitation of the tiny solid particles is avoided. It is essential to remember that it is impossible to obtain slips made of sufficiently resistant glass upon which water will not exert a marked solvent action ; moreover, it must constantly be borne in mind that all liquids soon take up foreign matter from the bottles in which they are kept. The results of tests for solubility should always be checked by comparison with the residues left when the solvent alone is evaporated under exactly the same conditions. It follows therefore that tests for the solubility of substances in boiling liquids or in strong acids, alkalies, etc., should be per- formed on clean, bright platinum foil; the solvent is decanted, concentrated and only transferred to a glass or quartz slip when evaporated almost to dryness. Should the illuminating gas be of very poor quality and the heating prolonged, an amount of various ammoniacal, sulphur and other products may be absorbed by the solvent sufficient to vitiate the results. If the substance whose solubility is being tested is subse- quently to be analyzed, a sufficient quantity of it is tested on glass, quartz or platinum, according to the necessities of the case, care being taken to observe the precautions given above as to impurities in solvents and the probability of their action on the microscopic slides used. This action may not always be due to the solvents alone, but may be the result of the material being tested. When more than one solvent has been found, the choice will, of course, be governed by many circumstances. It is obvious that no fixed rule may be given which will apply to even a majority of cases. Much must always be left to the judgment of the analyst. Decantation. For most purposes, it is generally possible to obtain sufficiently clear solutions from drops containing precipi- METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 231 tates or fragments by drawing off the supernatant liquid, without being obliged to resort to the longer and more tedious methods of nitration. Success in drawing off a liquid requires, in the first place, a perfectly clean slide free from grease, otherwise the liquid will not flow properly; and, secondly, patience, care and a steady hand. The first requirement is met by treating the slides in one of the usual cleaning mixtures of which the chromic- sulphuric acid is the best, and subsequently thoroughly washing them. Sometimes placing a drop or two of ammonium hydroxide on the slide and wiping it dry with a clean cloth will materially improve the surface. The other requisites for successful decan- tation are dependent upon the manipulative ability of the analyst and may be acquired only by practice. Although the phrase synonymous with decantation draw- ing-off is self-explanatory and the method is quite obvious, there are, nevertheless, several points upon which the success of the operation depends. Assuming that the drop of liquid is situated, as usual, at the corner of the slide, the operator proceeds as follows : The slide is held in a horizontal position ; the end of a drawn-out glass rod or a platinum wire is carefully introduced into the edge of the drop and is then slowly drawn across the slide (the slide being simultaneously slightly inclined in the same direction) until a distance of about one centimeter is reached. If the slide is per- fectly clean the liquid will follow the rod or wire in a narrow stream. A circular motion is now given the rod, resulting in the spreading out of the little stream into a drop ; this induces a flow of the liquid from the original drop. The steps in the decanta- tion are indicated in Fig. 123. The flow is aided by increasing the angle of inclination of the slide, providing, of course, there is no tendency on the part of the sediment to flow with the liquid. The important points, which can be learned only by practice, are the proper angle and the rate and manner of spreading out the drop. Should there be any tendency of the sediment to pass over with the liquid, reduce the angle at once. If the sediment tends to form a dam and prevent the passage of the clear liquid, it is neces- sary to start a new current at one side of the barrier or to break 232 ELEMENTARY CHEMICAL MICROSCOPY the latter down at a suitable point. As soon as the proper volume of liquid has been drawn off, still holding the slide in- clined, a piece of filter or folded lens paper is drawn through the channel, between the two drops at C, Fig. 123, and the prepara- tion immediately heated gently over the micro-flame at this same point. The result of this heating is the separation of the two drops by a dry space ; thus there is no danger of the decanted liquid flowing back when the slide is again placed in a horizontal position. FIG. 123. Decanting a Drop of Liquid from a Precipitate. When the clear decanted liquid is not wanted for analysis and only the sediment, or precipitate, in the original drop is to be utilized, the decanted portion and connecting stream are both wiped off the slide with filter paper while the slide is inclined and the preparation heated gently below the wiped-off drop to pre- vent any farther spreading. In cases where the sediment in the drop persists in flowing with the liquid being drawn off, and where heating is not objection- able, the slide is tipped so as to cause all the liquid to again flow back into the original source and the drop is evaporated to dry- ness at a low temperature, exceptional care being taken to pre- vent heating the residue after evaporation. This step will usually cause the sediment to cling to the glass and to aggluti- nate. A drop of water or the proper liquid is then -carefully added, the preparation allowed to stand a few seconds to permit the soluble compounds to pass into solution and the solution then decanted as above described. Usually a clear liquid may now be obtained without difficulty. METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 233 Liquids which have been decanted but which are not suffi- ciently clear may be evaporated and treated by the method described in the preceding paragraph. Washing precipitates by decantation may be performed by drawing off the liquid as above, adding a drop of washing liquid to the residue, allowing to stand for a few seconds and drawing off as before. The process is repeated as long as is thought necessary, or until tests applied to the decanted liquid prove that the washing is sufficiently complete. It is obvious that with a pure solvent, containing no compounds in solution, the simplest test is evaporation to dryness and the obtaining of no perceptible residue. In the event of a number of drops being obtained in the process of washing, all of which must be saved and united for subsequent examination, it is best to transfer them to a second clean slide; this is done by decanting into the extreme corner of the slide, cutting off the stream with filter paper and warming as already described. Now slowly raise the slide to an almost vertical position and bring the corner, holding the decanted drop, in contact with the slide prepared to receive it. Touch the drop at the corner with a drawn-out glass rod or platinum wire and the drop will flow at once on to the slide below. Raise the verti- cally held slide and warm its corner over the micro-flame, wash the residue as before and again transfer. The united washings may afterward be concentrated to the proper volume by evapo- ration. In all cases where decantation is to be practiced the size of the drop to be treated must be somewhat larger than that employed in tests alone. Decantation by Means of the Centrifuge. Next in impor- tance to the methods above described for separating sediment from liquid must be placed the centrifugal machine. A " two-speed" machine, with hematokrit frame, should be purchased, 1 since it is seldom that sufficient liquid is available in ordinary microchemical work to permit of the usual sedimen- tation tubes being employed. With the hematokrit attachment, 1 A convenient form of machine is shown in Fig. 76. 234 ELEMENTARY CHEMICAL MICROSCOPY however, very small quantities of liquids can be handled, and the high speed obtainable will throw out even a precipitate whose specific gravity differs but little from the liquid in which it is suspended. A convenient form of tube for use at high speeds may be made as follows: An ordinary glass tube of proper size is drawn out to a point in the flame of the blast lamp, and then, by continued heating, the glass is allowed to thicken a little at the end; the end is pressed, while still soft, against a piece of asbestos board, or a piece of charcoal, to flatten it sufficiently to fit well in the hematokrit frame. The tube is then cut the proper length, and the upper end smoothed with a file or rounded in the lamp flame. The turbid liquid to be treated is introduced into the tube by means of a pipette with long capillary end, and the tube is then placed in the frame; a similar tube is filled with water to the same height, and is placed in the other side as a balance. Thus arranged, the machine is turned at such speed and for such a time as may be necessary to yield a clear liquid. The treatment to which the sedimentation tube is then sub- jected will depend upon whether the liquid or the sediment (or both) is wanted. When the clear supernatant liquid is required, it is removed by means of a pipette with long capillary tip. But when the precipitate alone is needed the clear liquid is most con- veniently removed by capillary tubes, made by drawing out odds and ends of glass tubing. With such tubes it is only necessary to touch the liquid, which will immediately be drawn up by capillarity; the tubes filled as far as the force will raise the liquid are thrown away. One tube after another is inserted until the liquid is lowered to a point just above the sediment. Distilled water is introduced, and if the precipitate is to be washed, the contents of the tube are mixed well with a platinum wire, and the tube is again whirled to effect a separation; for most purposes one washing is sufficient. The wash water is removed as before, and if the amount of sediment is very small, the tube is cut off just above it to enable easy removal of the solid material. The upper part of the tube is not wasted, but serves to make capillary tubes. These small sedimentation tubes are easily and quickly METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 235 made. A stock should be provided so that a number are always on hand. It will be found convenient to have sedimenta- tion tubes of different diameters, to permit varying amounts of liquid being used. Similarly constructed smaller tubes of thinner wall can be made to fit inside the ordinary " sputum" tubes usually furnished with the centrifuge. Once having become accustomed to using this instrument, the worker in microchemistry will find that the two-speed centrifuge is an almost indispensable instrument, which will enable him to meet with ease all sorts of problems involving the separation of solids and liquids that would otherwise tax his patience and ingenuity. Especially to be recommended are electrically driven centri- fuges provided with protecting hoods. When dealing with relatively large volumes of liquid the usual conical sedimentation tubes, shown in Fig. 76, will prove useful, but since it is usually the sediment which is to be subjected to examination or analysis, and rarely the liquid, it will be found more convenient to employ tubes drawn down to a fairly long pointed end which may be cut off with a file scratch just above the sediment, thus permitting easy access to the solids thrown out from suspension. When properly drawn down, tubes of this form can be used several times by simply sealing the end; the tubes are centered and held in the aluminum carriers by means of perforated corks. Occasionally tubes with removable parts will be found to be convenient; the best forms are those devised by T. W. Richards 1 for the separation of small quantities of crystals from mother liquor. The construction and method of employment of these tubes will be readily understood by reference to Fig. 124. When one of the modern large electric laboratory centrifugal machines 2 is available very minute amounts of suspended matter may be separated from large volumes of liquid with great ease. The most convenient form of apparatus for this purpose con- sists in fitting a Squibb's separatory funnel with a stopcock of 1 Richards, J. Amer. Chem. Soc., 27 (1905) 104. 2 As for example the Bausch and Lomb Precision Centrifuge. 236 ELEMENTARY CHEMICAL MICROSCOPY the type provided in a Spaeth sedimentation glass, as shown in Fig. 125. Upon being whirled in the machine the suspended matter is forced into the conical cavity in the stopcock; a quarter turn of the stopcock completely cuts off the sediment from the Cork FIG. 124. Richards Tubes for Centrifugal Separations. FIG. 125. Sedimentation Fun- nel for Large Centrifugal Machines. liquid and the latter can be poured off without danger of disturb- ing the sediment; the stopcock can then be removed, and the contents of the cavity, containing only a very small volume of the solution and all the suspended matter originally present, subjected to examination and analysis. Filtration. In spite of every precaution it frequently happens that decantation will not yield a sufficiently clear liquid for sub- sequent reactions, or that the precipitate cannot be freed of the mother liquor, and that centrifugal separation cannot be used. Under such circumstances recourse must be had to nitration, which is doubtless one of the most troublesome processes of microchemical work. Since, in the majority of cases, the amount of liquid to be filtered consists of two or three small drops, often less, methods involving the use of a funnel, be it ever so small, METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 237 are to be regarded as unsatisfactory. In this category must be placed the ingenious filtering device of Haushofer, 1 for it is too cumbersome, complicated, requires too much time, and necessi- tates the transferring of the solution from the slide to the filtering apparatus, and back again to a slide. There are at present several practical and convenient methods for filtering small volumes of liquid, all based upon drawing the liquid through a tiny bit of filter paper held at one end by a glass tube of small or capillary bore while suction is applied at the other. The fundamental differences lie chiefly in the manner of applying the filtering material. None of these are to be recom- mended for qualitative analyses. The simplest, quickest and most useful method is that of Behrens. 2 A filtering tube is prepared, Fig. 126, consisting of a glass tube F about 60 millimeters long, and of 1.5 to 2 millimeters bore, with walls about i millimeter thick. One end is ground smooth and exactly at right angles to the axis; the other end is rounded so as to permit the easy attachment of a small piece of rubber tube R, about 80 millimeters long, carrying a piece of glass tube M for a mouthpiece. The preparation of the filter and the operation of filtering a liquid is performed as follows: A square piece of thick soft filter paper P of close texture is cut slightly larger than the diameter of the tube, and is placed on the slide S (which lies horizontally on the table) close to the drop D to be filtered; the ground end of the tube is pressed firmly against the filter paper near one edge ; the whole is then moved slowly into the drop; as soon as the paper is wet, gentle suction is applied to the upper end of the tube by the mouth, through the agency of the rubber tube. At the same time the filter paper is slowly advanced still further into the drop, the precipitate unless exceedingly fine will be pushed along in a ridge before the advancing paper and the liquid will rise in the tube. Care must now be taken to keep the rubber tube slightly curved, as shown in the cut. As soon as sufficient liquid has risen into the glass tube, suction is discontinued, the 1 Haushofer, Mikroskopische Reactionen, Braunschweig, 1885, p. 160. 2 Behrens, Anleitung Mikrochem. Anal., p. 22. 238 ELEMENTARY CHEMICAL MICROSCOPY rubber tube compressed at its upper end between the fingers and is simultaneously straightened to prevent the forcing out of the liquid. To lift the tube from the slide and the piece of filter paper, stretch the rubber tube very gently and raise the whole apparatus. The filtrate contained in the tube is removed by bringing the ground end in contact with a slide and bending the rubber tube, the upper end of which is kept closed; the liquid will generally flow out at once ; if not, straighten the tube, open the upper end and blow very gently, but only just sufficiently to expel the drops. FIG. 126. Behrens Method of Filtration. A little practice is required in order to apply the proper pressure of the glass tube upon the filter paper and to maintain this pressure uniformly without tipping the tube out of its vertical position. The chief difficulties encountered in rapid work are: (i) The danger of carrying the filtrate up into the mouth or into the rubber tube by air bubbles, which are always drawn into the tube when the liquid to be filtered has all been absorbed by the filter paper and sucked into the tube, and (2), it not infrequently happens that the filtered liquid begins to flow out when suction METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 239 is stopped and before there is time to prevent it by closing the upper end of the tube. These difficulties may be overcome by a modification of the simple filtering tube, 1 consisting of the intro- duction of an inner tube or trap. A glass tube about 3^ milli- meters internal diameter has fused into its vertical axis a tiny tube about i millimeter in diameter and 7 to 8 millimeters long. The lower end of the main tube is caused to flow together until the central opening is about 2 millimeters in diameter, and it is then ground so as to give a perfectly flat surface. The apparatus, which is 30 millimeters long, is attached to a rubber tube and is employed in the same manner as the previously described filter- tube. It is obvious that as the filtrate rises in the tube it over- flows into the small trap and is held in the space between the walls of the outer and inner tubes. The tube through which the liquid rises is therefore free, and any air bubbles entering cannot cause a loss of the filtrate, nor can the liquid flow back if suction is stopped. The filtrate can be removed either by means of a drawn-out pipette or by inverting the tube and in- ducing the liquid to flow by means of a platinum wire. Savage 2 introduces the filter paper within the tube, making the manipulation somewhat simpler, the filtering of liquids from very fine precipitates somewhat easier and permits of han- dling larger volumes of liquid. But this method fails to handle as tiny quantities of liquid as that of Behrens and the residue is not so readily separated from the filter. Savage describes his method as follows: "A glass tube of about 4 millimeters inside diameter is drawn out as abrupt as possible, and the narrow portion of the tube should extend from 15 to 30 millimeters from this point, with parallel sides and an inside diameter of about eight-tenths of a millimeter. The entire tube is 8 or 9 centimeters long, and both ends are rounded in the lamp flame. From a piece of soft filter paper of smooth surface and long fiber a triangular piece is torn (not cut), 2 to 2^ centimeters long and i centimeter wide at the base. This is rolled between the fingers into a slightly taper- 1 Chamot, Jour. Appl. Micros., 3, 854. 2 Savage, Jour. App. Micros., 3 (1900), 678. 240 ELEMENTARY CHEMICAL MICROSCOPY ing, cigar-shaped plug. It should be rolled dry and rolled long enough to make it fine and even. If the paper is cut, not torn, there will be a seam in it, and it cannot be so readily made tight. The plug thus formed is inserted in the small end of the tube from the outside and worked in by rotating the tube until from 4 to 8 millimeters of the paper are within. The rest of the paper is then cut off a millimeter or two from the end of the tube." The filter is first moistened with distilled water and then in- serted in the drop to be filtered, suction is applied to the larger end and the clear liquid drawn up through the filter into the tube, from which it is removed by a capillary pipette or by carefully removing the filter paper with a pair of fine forceps and expelling the liquid in exactly the same manner as in the Behrens method. A tightly rolled cigar-shaped plug of filter paper or fibrous asbestos may be inserted in a straight Behrens tube in a similar manner to that described above, and will be found to yield even more satisfactory results than the fragile drawn-out tube of Savage. The author has found in certain instances that alundum filters have proved of great value. Such filters are made by grinding tiny conical plugs from pieces of broken alundum crucibles and fusing these plugs into the ends of glass tubes 2 to 2.5 millimeters in diameter and 50 to 60 millimeters long. After fusing, excep- tional care must be taken in cooling and annealing. In like man- ner porous porcelain plugs may be used, but in such an event a powerful suction pump is required, suction by means of the mouth being insufficient to cause the passage of the liquid. Sublimation. This operation, though of somewhat limited application and comparatively seldom employed in inorganic qualitative analysis, is so very important, and of such inestimable value in the examination of organic compounds, that every worker should become thoroughly familiar with it, particularly with the method of performing fractional sublimations. The usual method is that of sublimation from one slide to another. The material to be tested is placed at the corner of a thin slide. If it is a solid it is wise to moisten it with water and then dry it thoroughly; this will generally effectually prevent METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 241 the material from being blown off by air currents, and brings the substance in intimate contact with the glass slide a matter of prime importance. If the material is already in solution, evapo- rate a tiny drop, but in this case it should not be spread out, as is commonly done with test drops. When the drop is dry, add another tiny drop on top of the residue left by the first; this in turn is dried, the process being repeated until, in the judgment of the operator, there is sufficient material for work. In all cases the residue to be treated should occupy but little space, yet should not be too thick, since, if fractional sublimation is to be practiced, a thick mass is apt to be heated unequally and fallacious results will be obtained. Everything being ready, the slide is held in the left hand and the heating begun over the micro-flame, not directly beneath the spot of material, but slightly nearer the center of the slide. This is done in order to avoid rais- ing the temperature too rapidly and too high. As soon as the sub- limation point is almost reached (which can easily be recognized by practice) a second clean slide, carrying a drop or two of water, is taken in the right hand and lowered over the first slip, with the drop of water on the upper side directly over the material to be sublimed. The drop of , . . , . , FIG. 127. Sublimation of Material water has for its object the from One Object Slide to Another. keeping of the upper slide cool, thus far more effectually condensing any vapors produced by the heating. The receiving slide is supported on an edge of the other and is brought to within 2 to 4 millimeters of the substance (see diagram, Fig. 127). The temperature is gradually raised by moving the spot of substance nearer the flame. As soon as there is evidence of the appearance of a sublimate, raise the two slides above the flame so as to prevent too rapid vaporization. The 242 ELEMENTARY CHEMICAL MICROSCOPY first deposit being obtained, the receiving slide is moved along a few millimeters and a second sublimation made; again the slides are partly removed from the source of heat, the receiving slide moved along a trifle, and again the temperature is raised until a third film has been condensed. The process is continued as long as the material holds out on the first slide or fails to yield any further sublimate. If the drops of water, used to keep the receiver cool, evaporate, replace them by others. When dealing with compounds which melt on heating, the supporting slide must be slightly inclined so as to keep the material at the corner of the slide. Or we may sublime from a watch glass upon an object slide, as shown in Fig. 129, page 245. It sometimes happens that a more crystalline and characteristic sublimation film is to be obtained when the receiving slide is slightly warm, in which event the water is omitted, or, if this is not sufficient, a little cylinder made of carbon, such as is used in arc lamps, is warmed over a burner and placed upon the slide. Such pieces of carbon remain warm for some time and will be found to give excellent results. With the beginner it is always best to obtain each fractional sublimate upon a separate slide, carefully laying them down film side up in the order in which they have been obtained. Other- wise the films first formed are apt to be driven off by the in- creasing heat required to vaporize the last portions or will be rubbed off by the fingers or by contact with the support. When a series of sublimation films are obtained upon a single slide always see that the films succeed each other in such a man- ner as to bring the first ones farther and farther from the source of heat as each film in turn is formed. When dealing with sublimations taking place only at tempera- tures so high that ordinary glass will soften, quartz slips may be employed or nickel or platinum foil or small nickel or platinum spatulas. The method of procedure will in any event be similar to that above described, intimate contact between substance and support being first accomplished when possible by moistening with water and careful drying. The temperatures of sublimation may be determined by means METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 243 of a hot stage such as that described on page 224 or by the method recommended by A. W. Blyth. 1 A small porcelain crucible is nearly filled with mercury, into which dips the bulb of a ther- mometer. A thin cover glass, bearing at its center the material to be tested, moistened and dried as usual, is floated on the surface of the mercury. Upon the cover glass is placed a low glass cell whose upper and lower rims are accurately ground. A second cover glass is placed above to receive the film see diagram, Fig. 128. A number of clean covers should be placed near at hand. The crucible is heated over the low flame of a Bunsen burner. As the temperature rises, the covers are changed, by means of a pair of forceps, every five or ten degrees. The cover glasses are examined under the microscope, and a decision made as to the temperature of FlG - I28 - Crucible -IT,. j i j.i j Method of Micro- subkmation. A second and even a third ex- , r . sublimation. periment should always be made. If the material fails to sublime at a temperature below that at which the mercury itself is volatilized, a bath of a suitable low-melting alloy must be used. For accurate measurements it is essential to protect the crucible and cell from the cooling effects of air currents. Subliming upon a glass object slide as shown in Fig. 127 is impracticable when only a minute quantity of the material is available since the losses through incomplete condensation are considerable. In such an event it is safer to employ the device shown in Fig. 130, page 245, primarily intended for distillation but yielding good results with solids as well as with liquids. When, however, only an excessively small amount of material is to be tested as in toxicological analysis, it is better to drop the substance into a thin-walled glass tube of not over i millimeter in diameter, sealed at one end. Tap the tube gently so as to col- lect all of the material at the sealed end. With a very fine blast- lamp flame draw out the open end to a hair-like capillary tube, 1 Poisons: Their Effects and Detection, 259, 4th Edition, London, 1906. 244 ELEMENTARY CHEMICAL MICROSCOPY and after cooling, gently heat the material in a hot stage of the type shown in Fig. 121, until sublimation takes place. The chief difficulty with the tube method lies in the fact that the poor quality of the glass, the striations, air bubbles, and defects render the examination of the sublimate complicated and diffi- cult. Laying the tube in a drop of oil or of glycerine at the point where the sublimate appears facilitates the study, by preventing the formation of heavy black contour bands. Distillation. Simple as well as fractional distillations are as important in the separation and identification of compounds in microchemical analysis as in the usual methods on a larger scale, and although one of the most difficult of microchemical methods may, nevertheless, with care and patience, be performed as successfully as the series of fractional distillations on the usual scale of the chemical laboratory. The simplest of the distillation problems arises in the detec- tion of a volatile constituent which can be expelled from non- volatile material by heating after the addition of a suitable reagent, as, for example, in the detection of ammonia by expulsion from material made alkaline with sodium hydroxide or in the detection of inorganic or organic acids set free from their salts by phosphoric acid and expelled by heat. The method of procedure is as follows: Place in a deep 25-millimeter watch glass a tiny bunch of fibrous asbestos which has just been ignited to redness by being held with the forceps in the flame of a Bunsen burner. In the absence of asbestos pure glass wool or in certain cases even a piece of filter paper may be employed as the absorbent, but if filter paper is employed a blank must always be made to prove that no misleading substances result. The asbestos or glass wool prevents the spurting and splashing of the liquid. Upon the absorbent is placed a small amount of the material to be tested, sufficient water and enough expelling reagent to just thoroughly moisten the mass but no more. Invert over the watch glass thus prepared a glass slide, bearing at its center a minute drop of water about i millimeter in diameter which has been acidulated or made alkaline as the case requires. Hold the watch glass thus covered by grasping its edges between the METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 245 thumb and forefinger, place a cooling drop of water upon the top of the slide and heat the watch glass gently over a micro-flame (Fig. 129) until vapors begin to con- dense upon the object slide. Heating to violent boiling must be avoided. The cooling drop upon the upper sur- face of the object slide* is removed, the slide raised from the watch glass and turned over with a quick move- ment. The proper reagents for dis- FlG - I2 9 closing the presence of the constituent being sought are added and the resulting preparation examined with the microscope. The method just described is applicable only to easily vola- tilized substances and where prolonged heating is unnecessary, but even in expelling ammonia, the fingers become uncomfort- ably hot. To avoid this discomfort the distilling device shown in FJgs. 130 and 131 may be employed. It consists of a tiny glass . Watch-glass Method of Distillation. FIG. 130. Apparatus for Microchemical Distillations. (Slightly Enlarged.) crucible C, whose upper edge is ground smooth and true, a sup- porting clamp made of spring brass wire W and an ordinary short object slide O. The component parts are shown in Fig. 131, and the apparatus in use in Fig. 130. Just as in the watch glass method fibrous asbestos or glass wool is employed as an absorbent, an acidulated or alkaline drop serves to retain the volatile constituent and a cooling drop is placed upon the upper surface of the condensing slide. A lever L serves to keep the 246 ELEMENTARY CHEMICAL MICROSCOPY clamp open when removing or changing the object slide serving as a cover. Instead of holding the watch glass and cover, at the edges, between the thumb and finger as described above, the clamp shown in Fig. 131 may be used, or two watch glasses with ground FIG. 131. edges selected to fit edge to edge may be clamped together. In certain instances either one of these watch glass methods may prove to be more practicable than the crucible. In all cases, however, the clamp support is far superior to the fingers. Although the device just described may be satisfactorily applied to the fractional distillation of small amounts of volatile liquids, small distilling tubes of the form suggested by Behrens 1 will be found in certain cases to be somewhat safer for very volatile substances. These are readily made from small glass tubing of thin wall as shown in Fig. 132; the different steps in the preparation are indicated in i, 2, 3, 4 and 5. The finished distilling tube is shown in A. To introduce the liquid to be dis- tilled into one of the tiny bulbs, fuse the end of one of the project- ing tubes, cool thoroughly and introduce the end of the open tube into the drop of liquid, warm the upper bulb to drive out air and allow to cool with the tube still dipping into the liquid. If an insufficient quantity of the liquid enters, heat again just enough to drive out all but a trace of liquid, then dip the end below the liquid to be tested and heat the lower bulb until the contents are 1 Anleitung z. mikrochem. Anal. (2 Auf.), p. 140. METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 247 vaporized; on cooling the bulb will fill with sufficient liquid. Now open the fused end of the tube and close that end of the tube through which the bulb was filled. Heat the empty bulb just enough to drive out any material which may have found its way therein. When absolutely cold, heat the liquid gently, distilling it over into the empty bulb. The distillate is expelled by carefully tipping the apparatus on its side and gently warming the air entrapped back of the distillate, or cut off the sealed end and blow gently. FIG. 132. Steps in the Preparation of a Behrens Distilling Tube. (Full Size.) A more universally applicable distilling tube is shown in Fig. 133. It consists essentially of a tiny tubulated retort with at- tached receiver. The liquid is introduced through the side arm which is then closed with a tiny plug of cork or rubber or by fusing. Upon heating the liquid the vapors pass down the nar- row inclined tube, are condensed and collect in the rounded receptacle. To prevent loss the narrow tube between retort and receiver may be wound with wet filter paper. The distillate 248 ELEMENTARY CHEMICAL MICROSCOPY is removed from time to time by means of capillary pipettes. This little apparatus also makes a convenient generator for hydro- gen and arsine in testing for arsenic. When temperatures of vaporization are needed the bulb con- taining the liquid can be introduced into the hot stage described FIG. 133. Tube for Microchemical Distillations. (Full Size.) on page 224, the receiving bulb being kept outside of the stage and cooled with wet filter paper, the tube connecting the two little bulbs having been bent at the proper angle. Ignition, Fusion, etc. Operations involving heating to red- ness are best performed in small platinum cups or spoons, Fig. 134, over the low flame of a Bunsen burner or that of a miniature blast lamp. FIG. 134. Platinum Cups for Fusions. (Full Size.) FIG. 135. Casserole for Microchemical Analysis. (Full Size.) In the absence of alkalies tiny cups with handles made of fused silica are convenient, Fig. 135; or tiny porcelain casseroles can be used. All the apparatus illustrated are standard commercial forms and may be obtained from dealers in chemical apparatus. Small crucibles are occasionally useful, especially those corre- sponding to No. 9 and 10 Meissan porcelain. Since, however, METHODS FOR HANDLING SMALL AMOUNTS OF MATERIAL 249 crucibles require a special support during ignitions casseroles will be found more convenient. Grinding, Crushing, Mixing. For grinding and crushing materials for analysis, the smallest available agate mortars are best. One not larger than 30 millimeters in diameter, Fig. 136, FIG. 136. Agate Mortar for Microchemical Analysis. (Full Size.) should be selected. It must be carefully scrutinized with a lens to see that its inner surface is properly polished and is free from fissures, pits and scratches. A mortar made from a first quality piece of agate, if properly cared for, should last a lifetime. CHAPTER XIII. THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS. In order that success may follow our efforts in the application of tests resulting in the production of characteristic microscopic crystals, it is essential that reagents be always applied in the best possible manner and in concentrations and under conditions such as will lead to the separation of a solid crystalline phase in a very short period of time. It is. therefore necessary that we first ascertain the best method of procedure for each particular re- agent. Most of the failures to obtain satisfactory results when attempting microchemical reactions are due to a lack of apprecia- tion of the importance of this fact. Manuals of microchemical analysis usually neglect to state definitely the best manner of adding a reagent to a drop to be tested, assuming that the in- vestigator will ascertain for himself the conditions which will yield him products most easily identified. Under similar conditions as to concentration, acidity and manner of reagent application, the crystalline phase will not only almost invariably separate with the same habit, but the crystals will usually develop to the same size and will lie upon the object slide in each experiment in the same positions with respect to faces. The following methods for performing microchemical reactions involve different manipulations and can be considered as typical procedures, each applicable to the detection of a number of dif- ferent elements or compounds. The student should perform them until he is sufficiently proficient to invariably obtain an unequivocal test and one yielding each time similar crystals of a similar size. The more insoluble the compound, the more rapidly the crystals will separate and the smaller they will be. For convenience for future reference these methods are here numbered and described in detail. 250 THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 251 /. A drop of a solution of the reagent is allowed to flow into a drop of the solution of the material to be tested. This method of applying the reagent is more often employed than any other, and is generally far preferable to the ad- dition of a drop of reagent directly to the solution to be tested. A perfectly clean object slide is required. Upon it near a cor- ner place a small drop of the solution of the material to be tested. This drop should be spread out until it attains a diameter of approximately 5 millimeters and a depth of not over half a milli- meter. A drop of the reagent of the same diameter but about twice the depth is next placed adjacent to the first drop at a dis- tance of 2 to 3 millimeters. The concentration of the reagent drop should usually be slightly greater than that of the substance being tested. By means of a platinum wire or drawn-out glass rod, a tiny channel is made to flow from the reagent into the test drop, the object slide being tipped very slightly to facilitate the flow, but under no condition should the two drops merge completely. Having a higher concentration in the reagent drop usually leads to a flow of this liquid at a lower level and therefore close to the object slide because of a slightly greater density than that of the solution of the sub- stance. Crystals thus tend to form upon the slide instead of floating about in the liquid. The more perfect crystal faces are on the upper side, or, in other words, that side most easily studied by means of the microscope. Crystals which float about usually grow downwards from the upper surface of the test drop and therefore have the well-de- IG ' I37 ' veloped faces on their under side, which must remain more or less invisible. The maximum sizes of drops are shown in the diagram, Fig. 137. The reagent drop R has been made to flow into the drop to be tested S through a tiny channel c. The crystalline phase constituting the identity test separates at p. 252 ELEMENTARY CHEMICAL MICROSCOPY EXPERIMENTS. a. Addition of Chloroplatinic Acid (platinum chloride) to a solution of a potassium salt (KC1). Application: testing for K, NH 4 , Rb, Cs, Na, many organic bases, etc. Repeat the experiment, using a fragment of CsCl in a drop of the same size as that of the potassium salt just employed. Note the instantaneous forma- tion of a precipitate and that crystals are very much smaller. Repeat again, using a very dilute solution of CsCl. Next try a solution drop of KC1 containing very little CsCl. Allow to evaporate spontaneously after the addition of the reagent, Cs separates first, then K. b. Addition of Ammonium Mercuric Sulphocyanate to a dilute solution of a copper salt. c. Addition of a solution of a Tartrate to a solution of CaCl 2 acidulated with HC 2 H 3 2 . II. The substance to be tested is added to a drop of the reagent. This method of applying tests is the one least often employed. It will prove successful in such reactions as require for the sepa- ration and characteristic development of the crystalline phase a constant addition of one component, in this case that to be tested for in small but almost uniform amount. The fragment of material is added to the center of a shallow broad drop. Warming gently will accelerate the separation of crystals. EXPERIMENTS. a. To a drop of a solution of Bi 2 (SO 4 ) 3 containing a trace of free HNO 3 , add a fragment of K 2 SO4. Applications Testing for K, for Na, for Bi, etc. ///. A tiny fragment of the solid reagent is added to a drop of the solution of the substance to be tested. This case is substantially similar to Method II, and is governed by the same general conditions. It will be found to be the safest procedure in nearly all reactions where the solid phase at first formed is soluble in excess of the reagent, for there will always be during an appreciable time (owing to the rather slow solution of the reagent) a zone in which the equilibrium is such that the solid phase can exist. Thus the fragment of reagent will be surrounded by a clear space or ring, at the outer edge of which the solid THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 253 crystalline phase will easily be distinguished under the micro- scope. If the fragment of reagent added is too large, the clear ring rapidly increases in diameter as the reagent dissolves, and the solid phase is correspondingly rapidly forced toward the cir- cumference of the test drop and eventually disappears completely. The test drop should be somewhat deeper than usual and should cover a relatively small area. Reactions involving no re-solution of the crystals first sepa- rating require no such careful attention to equilibrium conditions, nor do they necessitate such constant observation under the microscope in order that the progress of the reaction may be followed. In this class fall the precipitations of one metal by another metal which is more electropositive. If, for example, we make use of the electrochemical series of Wilsmore-Ostwald, 1 it is found that the metallic elements are arranged thus: + <- Mg, Al, Mn, Zn, Cd, Fe, Tl, Co, Ni, Sn, Pb, (H), Cu, As, Bi, Sb, Hg, Ag, Pd, Pt, Au,-> -. Theoretically each element in this series is able to replace the elements below it in the series which are less electropositive. Since in many instances the metal displaced will separate in characteristic crystalline form, the addition of a tiny piece of Mg or of Al to a very slightly acidified drop may be made to yield a beautiful test for metals farther along in the series. This type of reaction is also of great value in effecting separations prior to the application of identity tests, or in the separation of elements which may interfere with future testing. A knowledge of the electrochemical series is absolutely essential in all analyses of alloys where tiny fragments are not completely dissolved since there will be solution of one or more components and the precipitation of others upon the surface of the undissolved material. Furthermore, a study of the above series will reveal at once the fact that the addition to a test drop of a reagent with reducing properties will in all likelihood be followed by the par- tial precipitation of any metals present which fall in the electro- negative end of the series. 1 Zeit. phys. Chem., 36 (1901) 92. 254 ELEMENTARY CHEMICAL MICROSCOPY /// A . A tiny drop of the reagent is added directly to the test drop at its center. This procedure is effective in all cases where the crystalline phase, which is wished, is not too slowly formed, has great crystal- lizing powers and forms a large molecule. It may be said, that, in a general way, the addition of a drop of the reagent directly to the drop to be tested is applicable to practically all micro- chemical reactions. But in many special cases the crystals separating are not as characteristic nor as constant in their habit as in other methods, nor does the reaction take place with suffi- cient rapidity. The direct addition of the reagent is also practiced when a heavy agglutinated precipitate results, which must subse- quently be freed from its supernatant liquid and then recrys- tallized. The most frequent cases where reagent drops are added are in acidification, alkalinization, neutralization; and in the addi- tion of some reagent whose purpose is to mitigate the dele- terious action of some compound present, as, for example, the addition of sodium or ammonium acetate to prevent a free mineral acid from interfering with a test. Usually, however, a fragment of the solid acetate is added rather than a drop of solution. Or we may add a drop of glycerine solution to retard the formation of certain crystals. EXPERIMENTS. a. To a drop of a dilute solution of HgCl 2 add a fragment of KI. Note the kind of crystals formed and their position with respect to the fragment of KI. After the fragment of KI has dissolved leaving a clear area, add to its center a tiny fragment of CuSO 4 ; the HgI 2 which has dissolved will be reprecipitated. b. To a drop of a very dilute solution of HAuCLj (chloroauric acid) add a tiny fragment of T1NO 3 . In this case the characteristic crystals consisting of TIAuCU 5 H 2 O (?) form upon the fragments of the reagent. c. To a drop of PbNO 3 solution add a tiny drop of a dilute solution of CuSO 4 . Stir. Add a fragment of Na(C 2 H 3 O2), stir until almost dissolved. Now add a fragment of KNO 2 and follow with a trace of dilute HC 2 H 3 O 2 . Tiny black cubes of the triple salt 2 (KNO 2 ) Cu(NO 2 ) 2 Pb(NO 2 ) 2 separate. d. To a drop of a solution of PbNO 3 add a tiny fragment of metallic mag- nesium. Try in like manner a number of elements in the electrochemical series. THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 255 IV. The reagent solution is drawn in a narrow channel across a dry film obtained by evaporating to dryness a solution of the sub- stance to be tested. Reactions requiring a nice adjustment of concentration or leading to the formation of moderately soluble compounds, thus entailing a considerable loss of time waiting for the formation of crystals, if much liquid were present, are always best performed on the dry residue. Residues for such reactions should consist of thin, uniform films of material and are to be obtained only when scrupulously clean slides are employed, when only a small amount of the substance is present and when care is taken to avoid heating too hot during the evaporation. Gentle heating and blowing on the warm drop will give the best results. Heating should be done at the corner of the object slide over the tiny flame of the micro-burner, tipping the object slide so as to cause the drop to flow toward the corner and holding above the flame in such a position that the tip of the flame is nearer the middle of the slide. This prevents the liquid from creeping and from spreading. It is usually advisable to examine this film under a low power to learn whether it is thin and uniform in character. In cases where a ridge of the solid material tends to form around the edge, as will be the case if too much substance has been used, it is advisable to remove this ridge by means of the platinum spatula (Fig. 68), using it shovel- wise. The reagent is dissolved in a tiny drop of water placed just beside the dried test drop, and is then drawn across the latter with a quick stroke of a glass rod with drawn-out end, care being taken to avoid rubbing the slide in leading the reagent across. To facilitate the flow, the slide should be inclined a trifle in the direction the liquid is being drawn. The solution should never spread over the entire film of substance, but should remain as a streak of liquid dividing the dry spot in half. When the liquid completely covers the residue, it is usually due to one or more of several causes : too thick a film ; a slide that is not clean; heating after the residue was dry and so detaching it from the glass; too much reagent, or the presence of excessively soluble compounds or those which refuse to adhere to the glass. 256 ELEMENTARY CHEMICAL MICROSCOPY EXPERIMENTS. a. Obtain a thin uniform film of NaCl as described above. b. Near the residue (2 to 3 millimeters) place a drop of distilled water; acidify the drop by touching with a drawn-out glass rod which has been dipped in dilute HC 2 H 3 O 2 ; introduce a tiny fragment of UO2(C 2 H 3 O 2 )2. Warm the drop gently to facilitate solution, but do not evaporate. Cool. By a single, rapid stroke of a glass rod or platinum wire, draw a streak or channel of the reagent across the center of the dry material. Place the preparation upon the stage of the micro- scope and search the edges of the streak of liquid at once. Tiny faintly yellow triangular and tetrahedral crystals of NaC 2 H 3 O 2 UO 2 (C 2 H3O 2 ) 2 will be seen. Analytical applications Na, Mg, U, acetates. V. Upon failure to obtain a decisive test owing to the unsatis- factory separation of crystals, the delicacy of the reaction can be increased through the addition of another reagent which will produce a less soluble salt of the same nature. The chemical reactions involved in the practical application of this method of increasing the delicacy of microchemical iden- tity-tests are among the most interesting and instructive with which we have to deal. To properly apply and interpret them or to devise new tests to meet special conditions requires, in inorganic chemistry, a good working knowledge of the Periodic System of Mendelejeff : while in the case of reactions in the field of organic chemistry success can only follow a profound knowl- edge of the chemical and physical properties of the compounds to be studied. Considering the method only from the viewpoint of inorganic analysis, the delicacy of a test can be increased by introducing into the test drop, in which no separation of a crystalline phase has taken place, a salt whose base will form a less soluble com- pound than that originally present. For example, suppose a test for the presence of chlorides is being made by means of platinum sulphate and a salt of potassium; with much chlorine, potassium chloroplatinate will separate, but if we obtain no crystals, we may add a little rubidium sulphate to the drop. Should this yield no result, it can be followed by a little cesium sulphate and finally carried to the limit by the introduction of a thallous salt. With the potassium salt the limit of the test is 7~ 4 milligrams of chlorine, but with thallium 4~ 6 milligrams (Beh- THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 257 rens). That is to say, that while we may obtain proof of the presence of an exceedingly minute amount of chlorine through the separation of crystals of thallous chloroplatinate, approxi- mately one hundred times as much chlorine must be present in order that it may be revealed as potassium chloroplatinate. This plan of producing a less soluble salt is, in general, to be preferred to that of causing the separation of a solid phase by forcing back the dissociation, by means of strong acids, salting- out, or other similar processes, since well-formed crystals result in the first case, but abnormal, atypical salts are apt to appear in the other cases. EXPERIMENTS. Repeat Experiment IIIc, page 254, gradually reducing the concentrations until no triple salt separates, then add a fragment of CsCl; the triple nitrite of Cs, Cu and Pb will appear. In a new preparation carry the dilution a little farther, so that the Cs salt does not appear at once. Add a fragment of T1NO 3 . The deli- cacy of the reaction will be approximately K:T1 as 3: i. VI. The reaction can be hastened and the delicacy of the test increased by exposure to alcohol vapors. It was stated under Method V, that it is rarely desirable to employ a reagent that will force back the dissociation; the reasons being that the addition of such a reagent causes a too rapid separation of a solid phase and there is a tendency towards the production of malformed, skeleton or exceedingly tiny crys- tals. When, however, the separation of a solid phase is acceler- ated by the gradual absorption of a vapor in the test drop, thus reducing the solubility by forcing back the dissociation very slowly, it requires only a little care to assure the separation of characteristic, well-formed crystals. Alcohol is exceptionally well fitted for use in all cases where a crystalline compound is less soluble in alcohol than in water. One of two methods will be found convenient. Place near the test drop a small piece of filter paper. Saturate the paper with a drop or two of alcohol, carefully avoiding the addition of more than the paper will absorb. Cover the drop and paper with a watch glass (Behrens) ; or place a piece of paper at the bottom 258 ELEMENTARY CHEMICAL MICROSCOPY of a crucible, preferably a tiny glass crucible as described on page 245, or in a small beaker. Saturate with alcohol and invert over the test drop. Owing to the difference in the vapor tensions, alcohol will be absorbed by the aqueous solution and the crystal- line phase will rapidly separate. Only a very short exposure is necessary. When dealing with very thin films or tiny drops where there is a tendency to evaporate to dryness, exposure to alcohol vapors is especially valuable. EXPERIMENTS. a. Prepare a large drop of a moderately concentrated solution of PbNO 3 . From this large drop take two small ones. Allow one of them to evaporate spon- taneously. Treat the other with alcohol vapor as described above. Note the difference in time required for the appearance of crystals. b. To a dilute solution of a calcium salt add a drop of dilute H 2 SO 4 by Method 7, page 251. Sheaves, bundles and isolated acicular crystals of CaSO^ 2 H 2 O will separate. Prepare a solution of the calcium salt so dilute that no CaSO4 appears after standing two or three minutes. Expose to alcohol vapors and note that characteristic crystals are soon visible. VII. The reagent is dissolved in alcohol and a drop -of the alco- holic solution is employed as in Method I. Although we are here dealing with a mode of applying the reagent already discussed, alcoholic solutions need special men- tion because of the care required in their application. The re- marks which follow are equally applicable to any other solvents or reagents of lower boiling point than water or of different sur- face tensions. There is always a marked tendency of the alcoholic reagent to spread over the whole object slide, carrying with it the drop of solution to be tested, or breaking the latter up into so many drop- lets as to render reliable observations impossible. Not infre- quently considerable skill is essential to prevent this dissipation of material. *When an alcoholic reagent must be added to a reagent drop, always have the drop at the corner of the slide, and tip the slide slightly before the alcohol solution is applied to the glass near the drop; as the reagent leaves the rod or pipette increase the THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 259 inclination of the slide at once so as to cause the reagent to flow toward the material to be tested. Counteract any tendency of the reagent to creep up by immediately increasing the inclina- tion to an almost vertical position. Often the preparation cannot be laid flat upon the stage be- cause of the instant spreading of the alcoholic solution. In such an event, the corner of the object slide holding the liquid is in- serted in the stage opening and may be held in place by another slide placed upon the stage, carrying a piece of " plasticine" against which the inclined slide is pressed. The preparation can then be examined with a low power, focusing each different area as it is brought into the field by means of the stage centering screws. Because of the difficulties involved in the study of inclined preparations it is always better to first evaporate to dryness the drop of material to be tested so as to obtain a broad thin film (see Method IV) and use a reagent solution made with as dilute alcohol as will yield the proper conditions required in the test. EXPERIMENTS. a. Obtain a thin film of KC1 at the corner of an object slide. Place near by a drop of an alcoholic solution of freshly prepared sodium bismuth thiosulphate. 1 Tip the slide slightly and draw the reagent across the dry film. Yellow monoclinic 2 crystals of potassium bismuth thiosulphate separate. The salt is believed to have the formula 3 (K 2 S 2 3 ) - Bi 2 (S 2 3 ) 3 2 H 2 0. It is readily soluble in water, almost insoluble in alcohol. 1 The reagent is prepared as follows: Place in a small watch glass (25 mm.) a small drop of dilute hydrochloric acid; add repeatedly minute amounts of basic bismuth nitrate, warming gently from time to time and stirring thoroughly, until a trace of the basic nitrate remains undissolved; now add a bare trace of hydro- chloric acid; just sufficient to dissolve the little residue of bismuth salt, but no more; then add to the preparation a tiny drop of water. A permanent precipitate of bismuthyl chloride should result. If the first drop of water does not produce a permanent precipitate, another drop must be added. To this latter turbid solu- tion a saturated solution of sodium thiosulphate is carefully added, with constant stirring, a tiny drop at a time, as long as any of the precipitate remains undis- solved. An excess of sodium thiosulphate is to be avoided. A perfectly clear, faintly yellowish solution should result. To this clear liquid add alcohol (95 per cent) drop-wise, until a permanent turbidity results, which is in turn cleared up. by the addition of a single drop of water. 2 Hiiysse, Zeit. anal. Chem., 39 (1900), 9. 2<5o ELEMENTARY CHEMICAL MICROSCOPY b. Prepare a film of KC1. Draw across it an alcoholic solution of picric acid CeHaCNOa^OH. Potassium picrate CeHaCNOa^OK is obtained in long acicular prisms of the orthorhombic system. Try in like manner, Na, NIL; and Cs chlorides. Try with Na 2 CO 3 . VIII. The reagent is incorporated into a fiber of silk, cotton, wool, or in a filament of guncotton and the prepared fiber dipped into the drop of solution to be tested. The development of the methods for testing by means of textile fibers into which are incorporated the reagents to be employed, is due to Emich x and to Donau. 2 That variety of fiber is chosen which has the highest adsorptive power for the specific reagent to be used, as, for example, silk for adsorbing litmus; wool or silk for turmeric; silk or cotton for .gold; guncotton for adsorbing zinc sulphide, etc. Two methods of applying the reagent fiber to the test drop are in vogue; one consists in laying the fiber across the drop of solu- tion so that about two-thirds of its length will be outside the drop. The liquid is drawn by the capillarity of the fiber so that it gradually flows over its whole length. The second method consists in rolling a bit of beeswax between the fingers until a tiny slender cone is obtained about 10 millimeters long by 2 or 3 millimeters in diameter. One end of the reagent fiber is attached to the apex of the wax cone and the base of the cone is gently pressed against an object slide. A very minute rounded drop of the solution to be tested is placed upon the slide about 5 milli- meters away from the base of the cone; the cone is then bent over until the free end of the fiber dips into the liquid. The preparation is next placed upon the stage of the microscope and the instrument focused upon the fiber just above the drop. Through capillarity the liquid is drawn upon the fiber and the reaction resulting is easily recognized. 1 Emich, Monats., 22 (1901), 670; 23 (1902), 76; Ann., 351 (1907), 426. J Donau, Monats., 26 (1904), 545; Ann., 351 (1907), 432. THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 261 APPLICATIONS OF THIS METHOD. Testing for acidity or alkalinity Litmus-silk Differentiating between strong mineral acids and organic acids Congo-red-silk Testing for boric acid, borates Turmeric- wool Group reagent for the heavy metals. . . . Guncotton-zinc-sulphide Test for gold .... Adsorption upon silk, reduction with stannous chloride For the methods for preparing the fibers, see page 271. EXPERIMENTS. a. Test a very dilute drop of an acidulated solution with blue litmus-silk. b. Test a dilute drop of alkaline solution with red litmus-silk. c. Place a drop of a dilute solution of borax upon an object slide, acidulate with dilute HC1. Dip into the drop from a wax cone a fiber of turmeric-wool. Allow to evaporate spontaneously to dryness. Examine the fiber under the microscope. It should have a brownish color. Lay the fiber upon a slide and moisten with a 10 to 15 per cent solution of NaOH. If borates were present the fiber turns a bluish or lavender color. d. Into a tiny drop of a solution containing Au, lay a fiber of purified raw silk, warm gently until evaporated to dryness; carefully avoid too high a temperature. The fiber turns yellow or red. Treat with a dilute solution of SnCl 2 containing a little tannic acid. A purple color results, due to the precipitation of metallic gold. The beautiful red color of the silk fiber before the reducing agent is added is due to colloidal gold; the agglutination of the colloidal particles by the SnCl 2 gives rise to larger particles which appear purple. IX. The delicacy of the test is increased by taking advantage of adsorption phenomena, or the test itself depends upon the adsorptive properties of a compound. Although reactions of this type are those most frequently employed in the differentiation of structures, tissues, cells and cell contents in biology, histology and pathology, through the use of differentiating stains or dyes, their applications in the chemical laboratory to the common problems of qualitative analysis are limited. The basis for selecting a reaction involving adsorption phe- nomena or solid solution is that the resulting reaction shall con- fer upon a practically colorless body a color of sufficient intensity to render it more easily discerned. Whenever, therefore, stain- 262 ELEMENTARY CtfEMICAL MICROSCOPY ing or coloring can be quickly and simply accomplished, advan- tage should at once be taken of the fact. As examples of qualitative tests which may be considered as falling under this method, the following may be cited: In testing for perchlorates, the addition of a permanganate will yield colored perchlorate crystals. Iodine and bromine are revealed by their coloring starch granules, or the presence of a compound setting free iodine from an iodide or from an iodate is ascertained by starch. Or, on the other hand, starch is easily differentiated from other substances by staining with an iodine solution. Most oil or fat globules may be stained by alkanin. Fullers earth affords a simple means of distinguishing between vegetable and aniline dyes and in a few cases between certain aniline dyes themselves. In the microchemical examinations of rock sections, aluminum hydroxide can be stained with congo red and gelatinous silica with malachite green tests which may be employed in testing for " weathering," etc. EXPERIMENTS. a. Next to a drop of a dilute solution of HC1O4 or NHUCIO*, place a drop of RbCl solution (or KC1, if no Rb is obtainable). Cause the Rb to flow into the perchlorate (Method 7). In a few seconds colorless, characteristic crystals of RbClO4 separate. Place a drop of dilute KMnO4 next to the preparation and cause it to flow into it. The crystals of RbClO 4 will become colored pink. The resulting compound is a solid solution (isomorphous mixture) of the permanganate in the perchlorate, due to adsorption. b. To a drop of a dilute KI solution add a few granules of potato or arrow-root starch. Stir. Examine under the microscope. Add at the center a very minute fragment of pure KNOa or NaNO2. Examine again. The starch granules should appear at the most only very slightly colored. Add a trace of very dilute HCaHsOg or H 2 SO 4 . The starch granules turn blue or purple, due to adsorption of liberated iodine. Repeat the experiment, substituting a bromide for the iodide and (NH4)2S2Og for the KNO 2 X. The reagent dissolved in a volatile solvent is spread in a film upon an object slide in such a manner as to yield a coating or varnish non-crystalline in character, and across this prepared surface a solution of the unknown material is drawn. THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 263 Behrens 1 has successfully used this procedure in testing for the alkaloid quinine. Although no other practical application of this method of testing has yet been made, its possibilities in organic analysis are great, and the principle upon which the test is based is exceedingly interesting, namely, inducing crystalliza- tion in an amorphous mass through the presence of a mother substance dissolved in a suitable solvent. EXPERIMENTS. a. Dissolve a little quinine in dilute H 2 SO 4 , add to the drop of solution a fragment of KI, stir until dissolved, then add a fragment of KNO 2 . Decant from the brown amorphous mass, wash the precipitate once with water and dissolve it in C 2 H 6 OH. The alcoholic solution is flowed over a clean previously warmed slide so as to cover it with a thin homogeneous varnish. Examine under the micro- scope and make certain there are no pleochroic crystals. Dissolve a little quinine sulphate in dilute HC^sO^ Draw a narrow streak of this solution across the varnished surface. Immediately highly pleochroic crystals of iodo-quinine sulphate (Herapathit) will separate. The student should satisfy himself that this is actually an excellent test for quinine, although quinine was employed in making the reagent. Try, for example, pure cinchonine, cinchonamine, etc., in HC 2 HaO 2 solution; no Herapathit crystals will be obtained. XI. Testing for the evolution of gas from a substance when treated with a reagent. Dissolve in hot freshly drawn distilled water such an amount of pure gelatin (one or two square millimeters of sheet gelatin) that the solution just jells on cooling. It is essential that this jelly shall not possess too high a setting power nor yet be so thin that considerable time is required for it to set after melting. The substance to be tested, if a solution, should be evaporated to dryness in a thin film, or if a solid, very finely powdered or spread out in a thin uniform layer. Upon the dry residue a small drop of the melted gelatin is caused to fall, is quickly spread in a thin layer, and the slide allowed to stand upon a cool metal surface until the gelatin sets. The preparation is then placed upon the stage of the microscope and is focused. Next to the jelly drop is placed the reagent whose effect is to be tested, 1 Anleitung, z. mikro. Anal. v. wichtigsten organ. Verbind. Heft III, 92. 264 ELEMENTARY CHEMICAL MICROSCOPY and by means of a glass rod, the reagent drop is caused to touch the jelly mass. The reagent slowly penetrating into the jelly attacks the substance. If a gas of relatively low solubility is generated tiny gas bubbles will appear in the gelatin. Applied as above described the test has a somewhat wider range of usefulness than if the reagent (acid) is dissolved in the gelatin, as suggested by Behrens. In the event that the gas set free by the reagent is very soluble in water, no gas bubbles will appear; in such an event the gela- tin may be made the carrier of some reagent upon which the gas will react and be thus made to reveal its presence. EXPERIMENTS. a. Evaporate a drop of Na2COa solution. Cover with gelatin, test with HC1. b. Place a little CaCOs on an object slide, cover and test as above. c. Test a little zinc dust in like manner. d. Test a cyanate in like manner, using H 2 SO 4 . XII. An amorphous precipitate is formed by the reagent and requires special treatment to induce crystallization. It has already been pointed out that in microchemical quali- tative analysis an amorphous precipitate is the least desirable form in which a substance may be separated for identification. Nevertheless, it often happens that such precipitates are obtained either accidentally or when it is more expedient to thus remove a substance in order to prevent it from interfering in subsequent testing for other substances. In qualitative analysis by means of microscopic methods two classes of amorphous precipitates are met with : (a) Those which require solution in a special solvent from which a crystalline compound eventually separates, and (b) those in which crystal- lization can be induced by inoculation with a trace of the same compound in a crystalline condition. Special mention is here made of the treatment of amorphous precipitates because in a number of instances treatment with hot concentrated sulphuric or hydrochloric acids must be resorted to in order to obtain recognizable compounds. THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 265 When a precipitate is to be recrystallized from hot concen- trated sulphuric acid, it must be placed or formed at the corner of the object slide and any supernatant aqueous solution decanted. A moderate sized drop of the concentrated acid is then placed upon the precipitate and the slide immediately inclined at an angle of at least 30 degrees, to prevent the acid from spreading. Heat from a tiny flame is then applied to the object slide just below the upper edge of the drop, and as the acid fumes off the flame is brought nearer and nearer to the corner. As soon as it appears that sufficient material has passed into solution, the preparation is removed from the flame and allowed to cool for a few seconds, while still held in an inclined position. The inclined slide is then tipped so as to cause a slow flow to the adjacent corner (see page 232, Decantation), thus decanting the clear acid from the remain- ing insoluble precipitate, the channel of flow is cut off with filter paper and the slide inclined until it is almost vertical, thus causing the clear drop of acid to gather at the very corner of the slide. This corner is then touched to a clean slide and through a touch with a glass rod or platinum wire the drop is made to flow from the inclined slide to the horizontal one. A small clear drop is thus obtained. This system of attack can be employed in all cases involving re-solution in strong reagents. Where constituents dissolving from the glass slide are objectionable platinum foil can be em- ployed, eventually transferring as above to a glass slide. The second case mentioned arises most often in the analysis of organic compounds, as, for example, in the separation of a free base from its salts by means of an alkali. Although the amorphous appearing material will eventually crystallize sponta- neously if given sufficient time, it is usually desirable to hasten the formation of typical crystals. This can be accomplished by taking upon a platinum needle the most minute fragment possible from a portion of the pure base believed to be present and drawing it through the amorphous mass, crushing it at the same time. Crystallization of the amorphous material is almost always immediately started and proceeds with great rapidity. 266 ELEMENTARY CHEMICAL MICROSCOPY EXPERIMENTS. a. Add (by Method /) to a drop of BaCl 2 solution a drop of dilute H 2 SO4, evaporate to cause agglutination of the BaSO 4 ; add a drop of water, warm gently. Decant. Recrystallize the residue from hot concentrated H 2 SO 4 as de- scribed above. Cool and breathe repeatedly upon the drop. Study the crystals as they form. b. Repeat, using PbNO 3 instead of BaCl 2 . c. Precipitate AgCl from a solution of AgNOs. Recrystallize from concen- trated HC1. XIII. The material to be analyzed is exposed to the action of vapors or gases, or a reagent is exposed to vapors or gases resulting from the action of some compound upon the material to be tested. Oxidation of loosely bound sulphur to sulphate can usually be accomplished by placing a drop of bromine in a watch glass or crucible (use the apparatus, Fig. 131, page 246), inverting the drop of a solution of the substance to be tested over the bromine, warming gently in the hood and allowing the preparation to stand for five or ten minutes in contact with the bromine vapors. In many instances, the substance need not even be in solution, but can be merely in suspension, provided it is in a finely divided condition. No specific directions are necessary other than the caution that the inverted drop must never be so large that there is danger of its dropping off the object slide. Never perform oxidations with bromine save in the hood at a distance from all microscopes. After exposure to the oxidizing vapors, the slide is removed, turned right side up, the excess of bromine expelled in the hood by gentle warming and the remaining drop tested for the pres- ence of sulphates. In testing for the presence of a gas, as, for example, hydrocyanic acid, the reagent (in this case silver nitrate solution) may be in- verted over the container in which the gas is liberated, watch glass, crucible or test tube, or in testing for arsenic through the generation of arsine, the gases may be conducted through a tiny capillary tube containing a minute crystal of silver nitrate. The distilling tube, Fig. 133, page 248, serves as an excellent generator for applying this modification of the Gutzeit test for arsenic (see Figs. 138 and 139). THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 267 In a similar manner traces of moisture (or water of hydration in tiny crystals) can easily be recognized by placing a minute quantity of dry powdered fuchsine in a capillary tube and causing the moist air to pass over it by heating. The change from the greenish black powder to crimson droplets is very striking. Numerous other examples might be given. EXPERIMENTS. a. Place in the crucible of the apparatus, Fig. 130, two or three fibers of asbestos, drop upon them a single drop of bromine (in the hood). Invert over the crucible a drop of a solution of a sulphide. Lower the clamp and warm gently in the hood, until the crucible is filled with bromine rapors. Allow to stand for about five minutes. During this period test a portion of the unoxidized ma- terial for sulphates as below. Lift off the object slide from the crucible, turn it drop side up and evaporate to dryness; add a drop of water to the cool residue, then a tiny drop of HNO 3 . Decant if not clear, and finally test for sulphates by adding a drop of Ca(C 2 H 3 O 2 )2. (Method /.) CaSO 4 2H 2 O separates in the form of radiating tufts or X's of monoclinic needles or thin prisms. b. Place in the glass crucible a dilute solution of KCN. Cover it with an object slide, carrying a small drop of AgNOs upon its under side. Raise the slide just enough to permit dropping in several small grains of primary sodium car- bonate (HNaC0 3 ). Cover tightly at once and allow to stand for five or ten minutes. If, after this interval, no cloudiness is visible, warm the crucible gently. Remove the slide and examine it with a ^ inch or 8 millimeter objective. AgCN appears as small colorless prisms with obliquely truncated ends. XIV. Methods involving fusing the material in a bead of borax, micro cosmic salt or other medium. Some of the very earliest attempts to employ the microscope for the detection of minute amounts of material were made in conjunction with the blowpipe analysis of minerals. It was found that many substances yielded characteristic crystals when fused in borax beads before the blowpipe at high tem- peratures. Although of questionable usefulness in systematic analysis, this method is of sufficient interest to the student to be well worthy of trial and study. 1 To obtain a loop wind a platinum wire twice around a glass 1 See Sorby, Chem. News, 19 (1869), 124; Wunder, J. f. prak. Chem., 109 (1870), 452; Emerson, Proc. Amer. Acad. Arts and Sci., 6, 476. 268 ELEMENTARY CHEMICAL MICROSCOPY rod 2 to 4 millimeters in diameter. Heat the wire red hot, dip into borax (or other substance) and heat until a clear glassy bead is obtained of from i to 2 millimeters thick. Cool. Examine under the microscope, using a low power to assure the absence of crystals. Heat and touch to the powdered material to be studied. Then very carefully heat the preparation in the flame of a Bunsen burner until the borax or phosphorus salt bead just begins to melt. Avoid heating to redness. Cool and examine with a 16-milli- meter objective. Heat again, and again place under the micro- scope, thus following any changes which may take place. Should a blast lamp be employed for the heating care must be observed to avoid too large *and too hot a flame. This method can be made to yield good results in testing for calcium and magnesium and also for silicon, zirconium, titanium and molybdenum. Colored bead reactions are also obtainable, as for example in testing for Co, Ni, Cr, Mn, etc. The general principle of the method is, however, much broader in its scope since it comprehends all cases where a crystalline phase will separate from a transparent molten mass which solidi- fies upon cooling. XV. Testing with ^Hydrofluoric Acid or Silico fluorides . These reagents are applied in one of the manners already de- scribed, usually by Methods /, ///, or /// A . Specific comment is necessary, however, because of the im- possibility of employing ordinary glass object slides and because of the great danger of permanently damaging the objectives through the corrosive action of hydrofluoric acid vapors. Before undertaking any tests in which hydrofluoric acid vapors will probably be present, remove all objectives from the nose- piece save the lowest power, and place all microscope accessories at such a distance from the preparation as to render them safe. Take a small cover glass, carefully add a tiny drop of pure glyc- erine to its center and bring the drop in contact with the lower lens of the objective and press gently until the drop spreads out into a thin film, holding the cover glass in place. This is done to THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 269 reduce the danger of corrosion of the lens by the acid vapor. If a considerable period of time is occupied in a series of tests, the cover glass should be removed at intervals and the objective thoroughly wiped off and cleaned with lens paper moistened with water, dried and a new cover glass and glycerine applied. It is always preferable to have a cheap objective set aside, especially for hydrofluoric acid work, so as not to run the risk of ruining an expensive lens. For supports upon which to perform the tests, celluloid slips will be found convenient. The chief difficulty arises when gently heating the preparation, to cause development of the crystal forms, since nitrocellulose is very inflammable. Slips of cellulose acetate are therefore far preferable but are at present not com- mercially obtainable. Glass object slides coated with a film of "zapon" varnish, allowed to dry, and a second coat applied, yield good results when carefully prepared, but require as great care in heating as cellu- loid slips. A better device consists in coating glass object slides with "Bakelite," and heating in an oven to the temperature directed by the Bakelite Company for the particular grade of "Bakelite" used. Slides thus coated can be warmed without danger and yield good results. Whenever a critical case arises involving the detection of minute amounts of silica, titanium or zirconium, etc., it is best to have recourse to cellulose nitrate or acetate slips so as to pre- clude the possibility of error due to pores or fissures in the var- nished surface of a glass slide. Decompositions by means of hydrofluoric acid are best per- formed upon small pieces of platinum foil or in the tiny platinum spoons shown in Fig. 134, page 248. Subsequently the material can be transferred to cellulose slips or varnished slides for study. In selecting slips made from cellulose compounds, only such pieces should be chosen as are not badly scratched and grooved, and which are as nearly colorless as possible. Deep yellow slips are not suitable since in testing for sodium or for silica we depend 270 ELEMENTARY CHEMICAL MICROSCOPY for identification upon the faint pink tint of sodium silicofluoride as well as upon its crystal form. The same caution holds good for "Bakelite" varnish obtain one not highly colored if pos- sible and coat the glass slide with only a thin film. In coating glass slides with any protective varnish always carry the coating over the edges. Glass slides varnished with Canada balsam dissolved in chloro- form or xylene and subsequently dried in an oven at a slightly higher temperature than that of the room can also be used, but are not so convenient as the methods given above. Rathgen has recently called attention to an entirely different manner of employing fluorides in microchemical reactions. He has shown 1 that a very sensitive and characteristic reaction for aluminum may be obtained by mixing the finely powdered ma- terial with several times its weight of ammonium fluoride in a platinum cup or tiny platinum crucible, to which is then added four or five drops of sulphuric acid and the whole heated gently until all volatile fluorine compounds have been expelled; the heat is next slowly raised to drive off the sulphuric acid and the cup finally brought for a moment to a low red. After cooling, the residue is transferred to an object slide by means of a drop of water and a tiny brush. Aluminum gives tiny six-sided crystals and hexagonal plates. EXPERIMENTS. Experiments involving the use of fluorides will be found outlined in Chapter XIV under the elements Sodium, Barium and Silicon. PREPARATION OF SPECIAL REAGENTS. Ammonium Mercuric Sulphocyanate. To a concentrated solution of mercuric nitrate add a concentrated solution of am- monium sulphocyanate as long as a precipitate is produced. Filter; wash thoroughly the mercuric sulphocyanate obtained and transfer while wet to a beaker, and dissolve it by adding slowly drop by drop a saturated solution of ammonium sulpho- 1 Zeit. anal. Chem., 53 (1914), 33. THE METHODS OF MICROCHEMICAL QUALITATIVE ANALYSIS 271 cyanate; as soon as the precipitate has dissolved add a very slight excess of the ammonium salt. Too much ammonium sulpho- cyanate decreases the sensitiveness of the reagent. Litmus-Silk-Fibers. 1 Boil a little raw silk with water, decant and wash. Pour over the silk fibers a solution of pure litmus. 2 Evaporate on the water bath to small bulk, but not to dryness. Wash the colored silk with water and dry by pressing between filter paper and then allowing to stand in the air for a time. For red-litmus-silk, treat with the weakest possible acetic acid and dry by pressing between filters just before using. For blue-litmus-silk, treat with the weakest possible sodium hydroxide and dry between filters. Turmeric-Linen-Fibers. 3 Boil commercial turmeric root with twice its weight of alcohol until all the coloring matter is extracted. Filter and evaporate to dryness on the water bath. Dissolve the residue in an amount of dilute alcohol (to which a drop of sodium hydroxide has been added) equal to about half the weight of turmeric root taken. Place the fibers, previously boiled in water, in this solution and evaporate almost to dryness upon the water bath. Wash once with water, press between filter paper and treat with water acidulated with sulphuric acid. Press between filters, wash with water and dry. Zinc-Sulphide-Fibers. 4 Soak guncotton in a strong solution of zinc sulphate, decant the solution and pour upon the wet cotton a dilute solution of sodium sulphide. After standing a short time, decant, wash the fibers with water and dry them for use. Congo-Red-Silk. Prepare a strong solution of congo red, make it very slightly alkaline and heat it to boiling; while it is still hot stir in the coarse raw silk fibers and allow the prepara- tion to stand for a short time. Remove the fibers, wash very thoroughly with distilled water, press between filter paper and 1 Emich, Monats., 22 (1901), 670. 2 Wartha, Zeit. anal. Chem., 15 (1876), 322. 3 Donau, Ann., 361 (1907), 426. 4 Donau, Ann., 361 (1907), 432. 272 ELEMENTARY CHEMICAL MICROSCOPY dry for use. Cotton fibers may be employed instead of silk, but are not so good. Mineral acids, even when in exceedingly dilute solutions, turn the colored fibers blue. Dilute organic acids usually do not alter the color. Alkalies produce no change, but will cause a blue (acid) fiber to turn red. CHAPTER XIV. CHARACTERISTIC MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS WHEN IN SIMPLE MIXTURES. The methods of applying reagents and of performing the neces- sary manipulations arising in qualitative analysis have already been discussed at length in Chapter XIII, as well as the applica- tion of the simple polarizing microscope to the differentiation of chemical compounds in Chapter VIII. In the directions which follow it is assumed that the student is thoroughly familiar with these topics. As an aid to the recognition of common salts which may be met with, there has been given under each element the crystal system to which its common salts are to be referred. This has been done in the hope that the student will learn to employ the polarizing micro- scope and come to appreciate its many advantages as an invalu- able aid and great saver of time and labor. In these tabulations the following abbreviations have been used: (I) Isometric; (H) Hexagonal; (T) Tetragonal; (O) Orthorhombic; (M) Monoclinic; (Tr) Triclinic; and the salts arranged in the order named. SODIUM. Crystal Forms and Optical Properties of Common Salts of Sodium. 1 A. ISOTROPIC. Isometric. Chlorate. 2 The alums (double sulphates of Na and Al, Fe, Cr) (I); chloride (I); bromide (I); iodide (I); 3 molybdate (I or O). 1 In the following tabulations the data given have largely been obtained from Groth's Chemical Crystallography. 2 NaClOa although belonging to the isometric system exhibits circular polari- zation in crystals. Its solution is inactive. 3 Nal forms hydrates optically active. 273 274 ELEMENTARY CHEMICAL MICROSCOPY B. ANISOTROPIC. Hexagonal. Nitrate (pseudo 0) ; normal phos- phate; potassium-sodium molybdate; silico- fluoride. 1 Tetragonal. Orthorhombic. lodate; nitrite; potassium-sodium tartrate; normal tartrate; primary phos- phate. Monoclinic. Acetate; secondary arsenate; borates, tetra and meta; carbonate; primary carbon- ate; chromate; ferrocyanide; 2 oxalate, ferric- sodium; secondary phosphate; ammonium- sodium acid phosphate; sulphate; primary sulphate; thiosulphate; zinc-sodium sul- phate. Tridinic. Bichromate; bi tartrate; primary oxa- late. DETECTION. A. By means of Uranyl Acetate. Apply test by Method 7F, page 255. Sodium yields with uranyl acetate small faintly yellow tetra- hedra, appearing black by transmitted light. The compound formed probably has the formula NaC 2 H 3 02 UC^^HsC^V The crystals are iso tropic belonging to the isometric system. Potassium, rubidium, cesium and ammonium yield long needles or slender prisms of the tetragonal system of greater solubility than the sodium compound and therefore not appear- ing until the preparation has evaporated almost to complete dry ness. Because of the high solubility of ammonium uranyl acetate, Schoorl 3 has suggested its use for detecting sodium instead of simple uranyl acetate. The test thus made is more sensitive, but lacks the convenience of the method given above in that no 1 Na 2 SiF 6 is said to be pseudohexagonal. 2 Na 4 Fe(CN) 6 12 H 2 O is pseudotetragonal. 3 Lenz u. Schoorl, Zeit. anal. Chem., 60 (1911), 263. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 275 indication of the probable presence of K, Rb, Cs or NH4, can be obtained at the same time Na is being searched for. In the presence of magnesium there will be obtained in addi- tion to the tetrahedra of the sodium double salt large monoclinic crystals of a triple salt NaC 2 H 3 2 . Mg(C 2 H30 2 ) 2 3 (UO 2 (C 2 H 3 2 ) 2 ) 9 H 2 0, taking the form of rhombs or appearing to be octahedra, dodeca- hedra or having a more or less triangular outline with incurving sides. When, however, the amount of sodium is very small with reference to that of magnesium, only the triple salt will appear. As might be expected any of the other elements in the magne- sium group in the Periodic System, Gl, Zn, Cd, can replace Mg in the triple salt. Precautions. Too much free acid interferes with the test a further reason for evaporation to dryness before applying the reagent. Much magnesium gives rise to a film of salts so hygroscopic that a dry film cannot be obtained unless the salts are first converted into sulphates by evaporation with a little dilute sulphuric acid. Members of the calcium group often cause trouble. If, there- fore, an unsatisfactory test for sodium is obtained and subse- quent testing reveals the presence of Ca, Sr or Ba, these ele- ments should be removed by precipitation with sulphuric acid, the solution filtered or decanted from the precipitate and the filtrate evaporated to dryness on platinum (why ?) and again tested for sodium. Any compounds present in the material to be tested which will yield an insoluble precipitate with uranyl acetate, as, for example, phosphates, will naturally seriously interfere with the test or may absolutely prevent the detection of Na. In such an event the amount of uranyl acetate employed must be slightly more than sufficient to satisfy all the PC>4 present and to unite with the sodium to form the double salt. Under these condi- tions this test becomes unsatisfactory as applied above since it 276 ELEMENTARY CHEMICAL MICROSCOPY requires too much time. It is then better to flood the dry film with reagent, allow a few seconds to elapse for the establish- ment of equilibrium and decant the clear solution from the pre- cipitate of uranyl phosphate. The decanted solution must then be allowed to evaporate spontaneously until crystallization sets in, or the evaporation may be hastened by gentle heating. This test for sodium is also apt to prove unsatisfactory in the presence of much potassium. To remove the latter add per- chloric acid in slight excess. Evaporate to dryness, moisten the residue with perchloric acid and again evaporate. Extract the residue with alcohol; potassium perchlorate is insoluble; so- dium perchlorate passes into solution (Schoorl). Evaporate the clear alcoholic extract to dryness and test for sodium. A further caution is necessary relative to the possible inter- ference of elements such as Fe, Mn, Ni and Co, which can form double acetates with uranyl acetate and thus reduce the amount of the reagent available to form the double sodium compound. EXPERIMENTS. Test for Na in a. NaCl, Na 2 SO 4 , HNa 2 PO 4 . b. NaKC4H 4 6 ; and in 3 (Na 2 C 2 O4) Fe 2 (C 2 O 4 ) 3 . c. A mixture of NaCl and MgSO 4 and of NaCl and MgCl 2 . d. A mixture of Na 2 SO 4 and ZnSO 4 . B. By means of Bismuth Sulphate. First convert the compound to sulphate by evaporations to dryness with sulphuric acid. Apply the bismuth sulphate by Method //, page 252. Immediately after the addition of the unknown to the reagent, gently warm the preparation over the micro-burner. Sodium bismuth sulphate 3Na2SO4 2612(804)3 separates in the form of colorless slender rods or prisms with almost rounded ends, uniting in crosses, X's, or more or less star-like radiating clumps. The crystals separating near the circumference of the drop are usually shorter, stouter and more prismatic, while those nearer the center are more rod-like. It is these rod-like crystals with parallel extinction which are the more characteristic and MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 277 unless these are obtained the conclusion that sodium is present is unwarranted. Potassium sulphate yields plates having a hexagonal or coffin- like outline or six-pointed stars and rosettes. When first formed these plates appear as circular disks but they rapidly acquire six sides or grow into rosettes. Ammonium, rubidium and cesium form similar hexagons and rosettes. When both sodium and potassium are present, the rod-like crystals of the sodium double salt and the hexagons of the potas- sium salt each appear, permitting a simultaneous detection of sodium and potassium. The addition of a very minute quantity of glycerine to the preparation before heating usually yields better crystals and more reliable results. Precautions. Tufts of fine radiating needles appearing greyish or brownish by transmitted light must not be regarded as indicating the presence of sodium; neither should stout prisms or elongated plates with forked or broken ends. It is always best to remove members of the calcium group by means of sulphuric acid before applying the bismuth sulphate test. Calcium is especially to be guarded against since calcium sulphate may assume forms which simulate tlje sodium double salt; for although the crystals CaSO4 2 H 2 O are monoclinic and usually lie in positions yielding oblique extinction, the extinction angle is small and unless care is exercised the student may credit them with parallel extinction. Free mineral acids (especially nitric) greatly retard the sepa- ration of sodium bismuth sulphate. In the absence of bismuth sulphate the reagent may be pre- pared as follows: At the corner of a slide place a drop of dilute sulphuric acid; add to this drop a little basic bismuth nitrate and stir until the bismuth salt has completely dissolved. Heat carefully until the water has been mostly expelled, and crystal- lization of the bismuth sulphate takes place; then add a rather large drop of water and a very minute drop of 'dilute nitric acid. 278 ELEMENTARY CHEMICAL MICROSCOPY Stir for a few moments. The reagent drop should now slowly clear up, and a perfectly clear solution should result. If, how- ever, the quantity of bismuth nitrate employed has been exces- sive, a residue remains; it is then necessary to decant the clear liquid. On another slide, or better on platinum foil, heat with dilute sulphuric acid a few particles of the substance to be tested. Drive off the excess of acid; cool and stir to provoke crystallization. If the drop refuses to crystallize, add more of the substance and heat again. A drop of the reagent prepared as above is placed at the corner of a slide, and to it is added, at the center, without stirring, a little of the moist mass of the material to be tested, taken from the platinum foil. Warm the preparation gently by holding it for a second or two about one centimeter above the micro-flame. Cool rapidly and examine at once. This reaction is more valuable for potassium than for sodium and constitutes one of the best microchemical tests for bismuth. EXPERIMENTS. Test for Na in NaCl; HNa 2 PO 4 ; in mixture of salts of Na and K and in mix- tures of salts of Na and Ca. C. By Means of Ammonium Silico fluoride. See precautions given under Method XV, page 268. To the drop of the neutral, or at the most only slightly acid solution of the material to be tested, add a fragment of ammo- nium silicofluoride. Allow to stand some time (but never upon the stage of the microscope] or hasten the reaction by gentle warming. Sodium silicofluoride NasSiFe separates in the form of six- sided plates or prisms belonging to the hexagonal (?) system. Unless the crystals are excessively thin they appear with trans- mitted light to have a very faint rosy tint. They polarize only feebly. The corresponding potassium salt of like formula is much more soluble, separates only from decidedly concentrated solu- tions, and crystallizes in small, colorless cubes, octahedra and -combinations of these two, or in dodecahedra (isometric). A MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 279 hexagonal or pseudo-hexagonal modification of potassium sili- cofluoride is also known but is formed only at low temperatures. There is no possible danger, therefore, of confusing sodium and potassium. It is well to remember, however, that undue de- velopment of the diagonally opposite faces of an octahedron yields a crystal giving an image hexagonal in outline. The color of the crystal and its action on polarized light should leave no room for doubt as to its identity. From very concentrated solutions, in addition to potassium, Li, Ca, Sr, Mg, Mn, Fe, etc., may possibly separate. Barium, if present, is always precipitated with sodium, form- ing barium silicofluoride BaSiFe, which cannot be confused with the sodium salt since the barium compound crystallizes in rods or fusiform crystals singly, in crosses or in irregular masses. Neither calcium nor strontium are precipitated by ammonium silicofluoride, but each salt is liable to separate from too concen- trated solutions. The calcium salt CaSiFe 2 H 2 O (monoclinic) forms spindle-shaped crystals, and though these are grouped in rosette-like masses, they are not to be mistaken for sodium. The magnesium salt MgSiFe 6 H 2 is so much more soluble than those above mentioned as to never separate save upon evaporation or from very concentrated solution. Its crystals are rhombohedra, polarize strongly and do not have a six-sided outline. The silicofluoride of iron is isomorphous with the magnesium salt. It is evident that if silicon is present in the material under examination, we can test for sodium and silicon in one operation by adding ammonium fluoride and then acidifying. A pre- cipitation of crystals resembling sodium silicofluoride would point to the presence of sodium and silicon, or an element be- having, under like conditions, similarly to silicon. Thus we have titanofluorides, zirconofluorides and stanofluorides from elements of the fourth group; and from the transitional ele- ments, glucinum in the second group and boron in the third, we may have glucinofluorides and borofluorides of sodium. Of these compounds the titanofluoride is known to be isomorphous with the silicofluoride of sodium. 280 ELEMENTARY CHEMICAL MICROSCOPY In the absence of ammonium silicofluoride, pure silicon dioxide and ammonium fluoride can be added to the acidified drop of the solution to be examined. Precautions. Neither ammonium silicofluoride nor ammonium fluoride should ever be employed without having first been tested for the presence of sodium. If the reagents are found to be impure, it is necessary to sublime them in a platinum crucible, or receive the sublimate on platinum foil held over the material heated in a platinum cup. In the presence of much calcium the crystals of sodium silico- fluoride may become distinct hexagonal prisms instead of hexag- onal plates, a fact which must be borne in mind when working with material of unknown composition. The silicofluoride test is one of the most valuable at our com- mand in testing silicates for sodium, in which case we need add only hydrofluoric acid or ammonium fluoride and sulphuric acid. The addition of sodium and a fluoride gives us a test for Si, Ti or B. Remember that glass slides cannot be used in this test for sodium; that only low-power (i inch) objectives of great work- ing distance should be employed, and even then the front lens should always be protected in some way, as, for example, with a small cover glass held in place with glycerine, oil or other suitable substance. The preparation should be examined as rapidly as possible, and must be quickly removed from the stage. When the microscope is provided with a nosepiece, it is advisable to remove the objectives not in use before examining any prepa- rations liable to give off hydrofluoric acid or volatile fluorine compounds. The objective must always be thoroughly cleaned after any such tests. EXPERIMENTS. a. Test, as directed above, salts of Na in both neutral and acid solutions. b. In order to better appreciate the reasons for employing celluloid slips, place a drop of water on a glass slide, acidulate (but add no Na), then add the reagents and examine the preparation. c. Try to obtain crystals of K 2 SiF 6 from KC1. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 281 d. Add a little CaCla to a solution containing Na and test as above. e. To a solution of NaCl add a little SiO2 or a trace of sodium silicate, then add NH 4 F and an acid. /. Repeat using some Ti compound in place of that of Si. g. Test a salt of Ba as above, then a mixture of Ba and Na. Note that it constitutes an excellent test for Ba even in the presence of Na. POTASSIUM. Crystal Forms and Optical Properties of Common Salts of Potassium. A. ISOTROPIC. The alums (I); chloride (I); bromide (I); iodide (I); cyanide (I); molybdate (I); silicofluor- ide (I or H). B. ANISOTROPIC. Hexagonal. Barium-potassium ferrocyanide; bo- rate, tetra; silicofluoride (H or I). Tetragonal. Arsenate; cyanate; secondary phos- phate. Orthorhombic. Antimonyl tartrate; chlorate; chro- mate; nitrate; perchlorate; permanganate; sulphate; primary sulphate; sulphocyanate; primary tartrate ; sodium-potassium tartrate. Monoclinic. Carbonate; chlorate; ferricyanide; ferrocyanide; iodate; oxalates; normal tar- . trate. Triclinic. Bichromate; persulphate. DETECTION. A. By Means of Chloroplatinic Acid. Apply the reagent by Method /, page 251. In a few moments, relatively large and beautifully formed, strongly refractive, bright, deep yellow crystals of K 2 PtCl 6 appear. The usual form is that of the regular octahedron, some- times showing faces of the cube. Horizontally elongated octa- hedra, or octahedra shortened parallel to one of the pairs of faces, are not unusual. 282 ELEMENTARY CHEMICAL MICROSCOPY Since the crystals usually lie on one of the faces of the octa- hedron, there is apt to result an abnormal development of this face and the diagonally opposite and parallel face; the resulting crystal will thus exhibit an hexagonal outline when seen through the microscope, i.e., viewed from above. Combinations of cube and octahedron may lead to a somewhat similar appearance. Not infrequently preparations are obtained in which twinning is very marked, and others in which there is a grouping of crys- tals in threes or fours. Of the twin crystals, one form seems to predominate; it results from the union, in reversed position, of two halves of an octahedron where the dividing plane is parallel to the two opposite faces. The size and rate of development of the crystals formed will depend largely upon the concentration of the test drop. In very concentrated solutions, minute crystalline grains or the skeletons of octahedra are produced. In very dilute solutions the crystals appear only after some time. In case the test drop proves to be of the latter sort, heat it gently to cause slight evaporation, or expose to alcohol vapor, see Method F7, page 257- Thin crystals are lemon yellow in color, but those which attain a considerable thickness are of a decided orange tint. The best results are obtained from neutral solutions or those which are very slightly acid with hydrochloric acid. Excess of mineral acids is to be avoided, sulphuric acid in particular. Either evaporate and remove them, or mitigate their action by adding sodium acetate or sodium carbonate. If the latter salt is used, care should be taken to avoid making an alkaline solu- tion and a large excess of the chloroplatinic acid must always be used. Ammonium, rubidium, cesium and thallous-thallium also give 'octahedral crystals with chloroplatinic acid, the composition of the salts being similar to that of the potassium salt. The solubility of these compounds, and consequently the size of the crystals produced, decreases rapidly in the order in which the elements are named. Ammonium will give octahedra of the same size as those of potassium, hence its absence must be MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 283 assured before the test can be considered conclusive of the pres- ence of potassium. Salts of sodium form sodium chloroplatinate NaaPtCle 6 H 2 0, a quite soluble salt crystallizing in yellow triclinic prisms, having an extinction angle of about 22 degrees, and usually exhibiting brilliant polarization colors. It is seldom that well-formed, distinct crystals can be obtained, the result generally being an aggregate of imperfectly developed crystals. The salt is soluble in even strong alcohol, so that the addition of this reagent will not cause the separation of crystals, but evaporation is hastened. The chloroplatinates of potassium, rubidium, cesium and am- monium are isometric. That of glucinum, which is also obtained when evaporation is practiced, is tetragonal. Lithium forms a very soluble chloroplatinate similar to that of sodium. Precautions. If salts of ammonium are present, or suspected of being present, place a little of the material to be tested on platinum foil, moisten with water, dry and ignite carefully, until all the am- monium salts have been driven off. Dissolve a portion of the residue in water, with the addition of a little hydrochloric acid if necessary; transfer to a glass slide, and test; then again ignite the remainder of the residue and test again. The reagent should never be employed, even though freshly prepared, without first testing it by evaporation to ascertain whether octahedral crystals are deposited, since potassium may have been extracted from the containing vessel, or ammonium absorbed from the air. In making the reagent from metallic platinum it must be borne in mind that the acids employed may contain salts of potassium or ammonium, or both. When the potassium salt consists of a compound other than the chloride it is always best to evaporate repeatedly with strong hydrochloric acid before applying the platinum reagent. EXPERIMENTS. a. Test as above KC1, NaCl, NI^Cl. b. Test a phosphate, a sulphate, and a tartrate of potassium. c. Test K 2 SO 4 in the presence of much H 2 SO4. 284 ELEMENTARY CHEMICAL MICROSCOPY B. By Means of Bismuth Sulphate. For method of applying the test and discussion of the prop- erties of the salt formed see Test B under Sodium, page 276. Potassium bismuth sulphate 3 K 2 SO 4 Bi 2 (SO 4 )3 separates first as circular disks which later develop into hexagonal plates or the skeletons of hexagons, i.e., six-pointed stars and rosettes. Ammonium salts yield similar crystals. Hence this test can- not be used to differentiate between potassium and ammonium. Precautions. See Sodium, Method B. EXPERIMENTS. See Sodium, Method B. C. By Means of Perchloric Acid. Apply the reagent by Method 7, page 251. In a few seconds, colorless, highly refractive, clear-cut crys- tals of potassium perchlorate KCICX separate. These crystals belong to the orthorhombic system, but at first sight those first formed usually appear to be isometric, while later, forms which might be mistaken for monoclinic prisms appear. Rubidium and cesium give a like reaction, and their per- chlorates are more insoluble than that of potassium. Thallium forms an even more insoluble perchlorate. The per chlorates of the elements of the other groups that are generally met with in ordinary work, are sufficiently soluble not to interfere. Potassium, rubidium, and cesium perchlorates possess a re- markable adsorptive power for potassium permanganate. The crystals are not altered in habit, size or rapidity of formation but become colored rose or rose-violet. The compounds resulting are a solid solution of potassium permanganate in the per- chlorates and are considered by crystallographers to be iso- morphous mixtures of the two salts. Advantage may be taken of this property of the potassium salt to obtain an exceedingly beautiful test, for if the test drop contains sodium permanganate, the potassium perchlorate sepa- rating therefrom will be colored. Add to the test drop a little MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 285 sodium manganate, 1 so as to impart a distinct green, then add a tiny drop of hydrochloric acid, thus converting the manganate into permanganate. The perchloric acid is then caused to flow in. The crystals of potassium perchlorate which separate have the same form as before, but are a beautiful deep rose color, the color intensity varying with the amount of permanganate present. In a few moments the liquid is completely decolorized, and the precipitated crystals deeply colored. Performed in this way the test is a most interesting and instructive one. The perchlorate reaction is of more value for the detection of the acid by means of rubidium chloride and for the removal of potassium to prevent interferences with tests for other elements, than for the identification of potassium. Precautions. To obtain truly satisfactory results, careful attention to con- centrations must be given, for if the solution is too concentrated potassium perchlorate is precipitated at once in malformed or skeleton crystals; while if too dilute the separation of the solid phase is too slow. Exposure to alcohol vapor hastens the reaction. In the absence of perchloric acid ammonium perchlorate may be used. EXPERIMENTS. a. Try the above reaction with different salts of K. b. Introduce NaMnO 4 into the test drop, and test as above. c. Make a mixture of K and Na salts. Treat a drop of a solution of this mate- rial with HC1O 4 , evaporate, treat with the reagent again and again evaporate, extract the dry residue with alcohol, and test the alcoholic extract for sodium with U0 2 (C 2 H 3 2 ) 2 . d. Try the action of HC1O 4 on members of the magnesium group, and upon members of the calcium group. AMMONIUM. Crystal Forms and Optical Properties of Common Salts of Ammonium. 1 Sodium manganate is employed instead of sodium permanganate because it is more stable as a laboratory reagent. 286 ELEMENTARY CHEMICAL MICROSCOPY A. ISOTROPIC. The alums (I); chloride (I); bromide (I); iodide (I); silicofluoride (I). B. ANISOTROPIC. Hexagonal. Fluoride. Tetragonal. Borate (NH^I^OT 4 H 2 0; primary phosphate. Orthorhombic. Bicarbonate; nitrate; 1 primary oxalate; normal oxalate; perchlorate; pri- mary tar tr ate; sulphate. Monoclinic. Secondary arsenate; bichromate; chroma te; molybdate; persulphate; ammo- nium-sodium acid-phosphate; secondary phosphate; primary sulphate; ammonium- ferrous sulphate; sulphocyanate; normal tartrate; thiosulphate. Tridinic. DETECTION. Unless the analyst is dealing with a simple salt of ammonium, it is always best to expel the NH 3 from the compound by distilla- tion (see page 244) with sodium hydroxide or magnesium oxide. The ammonia set free is fixed by absorption in a drop of dilute hydrochloric acid (or other acid). The resulting solution of ammonium chloride is concentrated or evaporated to dryness and the material thus obtained tested for ammonium. A. By Means of Chloroplatinic Acid. See Method 7, page 251, and discussion and precautions given under Potassium, test A, page 281. B. Through the Formation of Ammonium Magnesium Phos- phate. The typical reaction for this identity test may be written + MgCl 2 + HNa2PO 4 + NaOH = NH 4 MgP0 4 + 3 NaCl + H 2 0. 1 NH 4 NO 3 is pseudotetragonal. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 287 To the drop to be tested add a fragment of sodium phosphate and a very little magnesium chloride, stir thoroughly. Beside the drop place a drop of dilute solution of sodium hydroxide and cause this drop to flow into the other. Ammonium magnesium phosphate separates in crystals having the formula NHLiMgPC^ 6 H 2 0, belonging to the orthorhombic system and exhibiting an exceptionally strong tendency to assume hemihedral, hemimorphic and skeletal forms. This compound usually separates first as an almost amorphous precipitate which soon changes into star-like and X-shaped crystallites. Soon the X's fill out and envelope-like crystals result and at the same time rectangular prisms resembling roofs of houses appear. In preparations containing but little of the ammonium mag- nesium phosphate the stars and X's are usually absent. Precautions. Since the amount of ammonia obtained upon distillation is usually small it is quite necessary to avoid an excess of the mag- nesium salt and also the phosphate, for the reason that magne- sium phosphate is almost sure to be precipitated. This latter salt appears as an amorphous deposit and if conditions are favor- able it may eventually crystallize in star-like crystal aggregates, distinct, it is true, from the ammonium magnesium phosphate, yet very apt to confuse the beginner. If the phosphate test be applied directly to a solution of the unknown salt it must be remembered that both phosphates and hydroxides of a number of elements will probably be precip- itated. EXPERIMENTS. Test as above for the presence of NEU in several different salts containing this radical. CALCIUM. Crystal Forms and Optical Properties of Common Salts of Calcium. A. ISOTROPIC. 288 ELEMENTARY CHEMICAL MICROSCOPY B. ANISOTROPIC. Hexagonal. Carbonate (H or 0) ; chloride. Tetragonal. Oxalate. Orthorhombic. Ar senate (0 or M); chroma te (0 or M); tartrate. Monodinic. Nitrate; sulphate; double sulphates of calcium and sodium or potassium. Triclinic. Ferrocyanide. DETECTION. A. By Means of Dilute Sulphuric Acid. Apply the reagent by Method /, page 251. If calcium is present, monoclinic crystals of calcium sulphate will rapidly appear near the circumference of the drop of the substance. These crystals take the form of exceedingly slender, colorless, transparent needles, either singly, in sheaves, in bundles or in star-like clusters. When in tiny sheaves near the edge of the drop the crystals have often a more or less brownish tint when seen by transmitted light. Shortly after the appear- ance of the bunches of needles at the periphery, long, thin, slender and plate-like prisms with obliquely truncated ends are formed throughout the drop. These prisms are frequently twinned, yielding so-called arrowhead or swallow-tailed and X-like twins. These twin crystals are the most characteristic of the forms assumed by calcium sulphate of the formula CaS0 4 2 H 2 0. If no crystals are visible after waiting a short time, the prepa- ration may be cautiously concentrated. This procedure (evapo- ration) may, however, lead to the separation of such an amount of other salts as to render difficult the detection of the crystals of calcium sulphate. A better plan is to hasten the separation of the calcium salt by exposing the test drop to the vapor of alcohol; see page 257, Method VI. Salts of strontium may, under exceptional conditions (if the preparation be examined at once), yield a precipitate which closely resembles that given by calcium. These crystals of strontium sulphate rapidly disintegrate, however, and there MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 289 results a fine granular deposit. This granular or sandy deposit is the form assumed by strontium sulphate under the conditions which ordinarily obtain in this test. Barium is immediately pre- cipitated in an exceedingly finely divided condition, amorphous in appearance, but occasionally BaSC>4 separates in crystalline form (see Barium). Any lead which may be present will also be precipitated as a dense white amorphous powder. Occasionally, however, lead will yield a precipitate consisting of orthorhombic crystals. Silver will separate as Ag 2 SO 4 in the form of colorless, highly refractive, orthorhombic prisms, rhombs or crystallites of char- acteristic appearance. When the drop of sulphuric acid flows into the drop to be tested which contains mercurous nitrate or other soluble mercu- rous salts, the mercurous sulphate produced often assumes at first the form of acicular needles, closely resembling those of calcium sulphate; they are, however, blackish by transmitted light and rapidly take the shape of rod-like prisms quite distinct from the prismatic forms of the calcium salt. Precautions. Before applying the sulphate test, add a drop of dilute hydro- chloric acid to assure the absence of lead, silver and mercurous salts. If a precipitate is formed decant. It is not always wise to conclude that calcium is present when crystals, which apparently resemble the star- and sheaf-like ag- gregates of calcium sulphate, separate at once on the addition of sulphuric acid, even if the crystals exhibit oblique extinction. It sometimes happens that other compounds, not calcium sul- phate, separate in forms not to be distinguished, at first sight, from the crystals of the calcium salt. Such instances are for- tunately very rare. Allowing the preparation to stand a few minutes will usually permit the crystals to develop and their appearance will then be such as to avoid error. If, however, the analyst is still in doubt he may proceed as follows : After allowing sufficient time for the separation of almost all the calcium as CaSO4 2 H 2 O, draw off the supernatant liquor, add to the residue 290 ELEMENTARY CHEMICAL MICROSCOPY a solution of ammonium carbonate, the crystals of calcium sulphate will be dissolved and highly refractive rhombs and grains of calcium carbonate will appear; these are easily found by examining the preparation between crossed nicols. A high power is generally required. A serious interference is that of the chlorides of the trivalent metals. In the presence of these salts in large amounts it is generally advisable to proceed thus: Add to the somewhat dilute solution, ammonium acetate, heat to boiling, but avoid long or violent ebullition, since in the latter case the precipitate formed often refuses to settle. The clear liquid is then sepa- rated from the precipitate (by dra wing-off on the slide, nitration, or by means of the centrifuge), concentrated if necessary, and tested for calcium with sulphuric acid. Behrens states that calcium cannot satisfactorily be detected in the presence of borates; this appears to be true when only a minute quantity of calcium is present with a high percentage of boron and other salts; in such an event test by Method B. Strong mineral acids, in excess, so increase the solubility of calcium sulphate as to require evaporation almost to complete dry ness before the crystals of this salt appear. The addition of a fragment or two of sodium acetate or of ammonium acetate is always necessary in such cases before the sulphuric acid drop is allowed to flow in. This method of mitigating the action of the free acids, also reduces the delicacy of the reaction because of the formation of more soluble double sulphates of calcium and sodium or ammonium. Hence the addition of an excess of a soluble sulphate instead of sulphuric acid is not to be recom- mended. EXPERIMENTS. a. Try reaction, in the manner given above, on salts of calcium in neutral solution. b. Try the effect of precipitating in the presence of free HC1; then in the presence of free HNO 3 . c. Precipitate with dilute H2SO4, then heat, adding more acid if necessary, until white fumes are given off, cool, breathe on the preparation and examine. Calcium will separate either as the salt CaSO 4 , or as CaSO 4 H 2 SO 4 . The crystal forms most frequently met with are thin, rounded, prism-like plates or fusiform MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 2QI crystals with tufted ends. This modification of the test is not satisfactory for Ca, but is characteristic for Ba and for Sr (q.v.). d. Try testing for a trace of Ca in the presence of a large quantity of salts of the elements of Group I. e. Try effect of a solution of (NH 4 ) 2 CO 3 on crystals of CaSO 4 2 H 2 O. B. By Means of Oxalic Acid. Apply the reagent according to Method /, page 251. The oxalate which separates at room temperature from neutral or slightly alkaline solution has the formula CaC 2 O 4 3 H 2 O, and belongs to the tetragonal system. The crystals are tiny, highly refractive octahedra, or rectangular or square plates. If rapidly formed, crosses and bundles or sheaves of crystallites will be seen. From hot or acid solutions a monoclinic oxalate CaC2O4 H 2 O separates which is practically valueless as an identity test for calcium. This same salt appears to sometimes, separate if a large excess of oxalic acid has been added. In addi- tion to changing the crystal form free mineral acids so increase the solubility of calcium oxalate as to sometimes prevent its. precipitation. Strontium gives with oxalic acid an identical reaction, save that the crystals of strontium oxalate are generally somewhat larger. Barium oxalate takes the form of fibrous bundles of needles, and is not likely to be mistaken for either calcium or strontium. Zinc under certain conditions may yield a zinc oxalate difficult to distinguish from the oxalates of calcium and strontium. Magnesium oxalate will separate in forms not to be distin- guished from calcium oxalate if the test drop contains much acetic acid, but in the absence of this acid magnesium oxalate will not appear. Manganese forms groups of radiating needles (see Manganese). Lead oxalate may also assume forms somewhat resembling those of calcium oxalate, but after a short time these crystals, grow into large, well-developed prisms. Silver oxalate separates first as a granular deposit, soon changing to crystals of a great variety of forms, hexagonal plates, six-sided plate-like prisms and stout prisms with obliquely truncated ends. 2Q2 ELEMENTARY CHEMICAL MICROSCOPY In the presence of stannic chloride Behrens has shown that calcium oxalate assumes the form of tiny oval grains exhibiting an octahedral tendency while strontium yields large clear-cut beautifully developed tetragonal octahedra and barium gives short stout prisms singly, in crosses and in radiating masses, or if much barium is present, fusiform crystals and bundles of radiating needles are seen. Precautions. Oxalic acid, under favorable conditions, can cause the separa- tion of oxalates of the following elements: Gl, Ca, Sr, Ba, Mg, Zn, Cd, Tl; rare earths; Sb, Bi, Sn, Pb, U, Mn, Fe, Ni, Co, Cu, Ag. In the event of a precipitate of doubtful composition being 'obtained, draw off the supernatant liquid, or separate by means 4 (or some- times probably SrSC^ H^SC^) . These tiny plates eventually develop into more or less irregular spindle-shaped crystals, which gradually enlarge at the middle until they become irregular crosses with two very short arms. The appearance is very characteristic. The only element liable to lead to error is lead which often first assumes forms closely resembling those of stron- tium, later growing into crystallites which may be mistaken for barium. Recrystallized from concentrated hydrochloric acid strontium sulphate has an entirely different habit. Square and rectangular plates appear followed by thin prisms and sheaves of slender pointed crystals. The solubility of strontium sulphate in hydro- chloric acid is quite low, hence it is necessary to employ a large drop of the solvent and therefore it is seldom that all the pre- cipitate will dissolve. It follows that to obtain the best results the solvent should be decanted from the precipitate immediately 294 ELEMENTARY CHEMICAL MICROSCOPY after heating, and before crystallization sets in. The resulting crystals are quite small and of varied form. The results are less satisfactory than with sulphuric acid, but there is, on the other hand, the advantage that barium sulphate is practically insoluble in hydrochloric acid. It is of course essential in re- crystallizing from hydrochloric acid that not more than mere traces of free sulphuric acid be present. Free nitric acid should be absent. Before any attempt is made to recrystallize the precipitate of strontium sulphate, it is advisable, and usually necessary, to remove any calcium which may be present. This is accom- plished by extracting the precipitate with hot water in which the calcium salt is soluble. Unless this is done, peculiar crystal forms are obtained which are difficult to interpret. If only a small amount of barium is present, characteristic crystals of strontium sulphate are still obtained from hot sul- phuric acid, but much barium is apt to alter the usual crystal form, although the appearance of the crystals separating still suggests the strontium sulphate type. An excess of barium seems to cause the majority of the crystals to assume forms somewhat resembling barium sulphate. But, in general, crystals of both strontium and barium sulphate can be distinguished in mixtures of these two elements. Any lead which may be present will be precipitated in an amorphous condition by the dilute acid, although under rare conditions it may appear crystalline. Recrystallized from hot sulphuric acid, the lead sulphate, as stated above, will separate in forms which at first closely resemble those of strontium sul- phate and which, later, grow to forms which may be mistaken for barium sulphate. Recrystallized from hydrochloric acid there is less danger of error. If in doubt, extract the precipi- tated sulphates with a solution of potassium or sodium hydrox- ide in which lead sulphate is soluble. Silver sulphate will appear as already described under calcium. Hence silver as well as most of the lead should first be removed with hydrochloric acid. As in the case of calcium, chlorides of the trivalent metals MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 295 and salts of boric acid may sometimes interfere with the forma- tion of typical crystals of strontium sulphate. EXPERIMENTS. a. To a drop of moderately dilute solution of SrCl 2 , add dilute H2SO4 and examine at once. b. Recrystallize SrSO 4 from H 2 SO 4 and from HCl. c. Try to recrystallize SrSO 4 from HCl in the presence of H 2 SO4. d. Make a mixture of Ca and Sr salts and add H 2 SO 4 . Recrystallize the product from H 2 SO 4 without having removed the Ca. In another portion remove the Ca by extracting with boiling water and then recrystallize the residue. B. By Means of Oxalic Acid. See directions given under calcium, Method B, page 291. The crystals of strontium oxalate are similar to those obtained with calcium, but are usually distinctly larger, and crosses, prisms, and four-pointed rosettes are more abundant and larger. The crystals are either tetragonal or monoclinic depending upon whether formed in the cold or separating from hot solutions. Precautions. To avoid error when testing with oxalic acid, it is always ad- visable, after the crystals have well formed, to draw off the supernatant solution and add dilute sulphuric acid to the pre- cipitate. If no crystals of calcium sulphate appear after a few minutes, add more acid and heat until white fumes appear, care- fully observing the usual precautions. Transfer the drop of acid to a clean slide, breathe on the drop and examine for fusiform crystals of strontium sulphate. EXPERIMENTS. a. Test a drop of SrCl 2 solution with H 2 C 2 O 4 . b. Treat the oxalate thus obtained with H 2 SO 4 and recrystallize. BARIUM. Crystal Forms and Optical Properties of Common Salts of Barium. A. ISOTROPIC. Nitrate (I). B. ANISOTROPIC. Hexagonal. Nitrite. 296 ELEMENTARY CHEMICAL MICROSCOPY Tetragonal. Orthorhombic. Chromate (O or M). Monoclinic. Chloride; chlorate; bromide; ferro- cyanide; acid-oxalate. Tridinic. Acetate. DETECTION. A. By Means of Sulphuric Acid. Read fully the directions and comments under Calcium and Strontium, pages 289 and 290, and 293 and 294. The amorphous or semicrystalline precipitate first obtained must be recrystallized from concentrated sulphuric acid before identification is possible. The recrystallized salt appears at first as tiny rectangular plates and X-like crystallites. In this stage of development it may be mistaken for strontium sulphate. Continue breathing upon the drop of acid; under the influence of the moisture absorbed the crystallites grow rapidly, still retaining their X-like shape but the arms of the X's become feathered. There is a marked tendency for two adjacent arms of the X to develop much more rapidly than the other two. These crystallites grow relatively large and are constant and peculiar to barium. In the presence of certain acids or acid salts, especially from hot solutions, crystallites of barium sulphate may sometimes be obtained immediately upon the addition of dilute sulphuric acid. In the event of a heavy precipitate being obtained with the reagent, it is wise to remove a small portion to another slide for crystallization, rather than attempt to dissolve the whole mass. Recrystallization in the presence of much calcium is to be avoided. First extract the calcium sulphate with hot water. In the presence of moderate amounts of strontium the crys- tallites of barium sulphate are generally not well formed. If strontium is in excess, the crystals separating from the hot sul- phuric acid have the general type of strontium sulphate, 'but are not well developed and exhibit an inclination to approach the X-forms of barium sulphate. For this reason it is advisable to remove any strontium which may be present by repeatedly MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 297 heating with hydrochloric acid, in which strontium sulphate is soluble, while the barium compound remains undissolved and can then be recrystallized by heating with sulphuric acid. Even in mixtures, however, it is almost invariably possible to find characteristic forms of both barium and strontium, providing the analyst has a little patience and carefully examines the entire preparation. Any lead sulphate which may be present will appear, first, in crystals very suggestive of strontium sulphate, then, in a short time, in larger crystallites which may at times be mistaken for barium sulphate. Treatment with hydrochloric acid, or, better, with sodium hydroxide, will remove the lead, leaving the barium salt unacted upon. Precautions. It is sometimes desirable to apply other tests to the precipitated sulphate in order to confirm the presence of barium. In such an event, transfer the washed precipitate to platinum foil or to a platinum cup and fuse with potassium carbonate. The fused mass is then extracted with water and the residue of barium carbonate dissolved in hydrochloric acid. This solution can then be tested for barium by any of the tests given below. Since chlorides of the trivalent metals sometimes interfere with the formation of characteristic crystals of barium sulphate, it is advisable to decant the supernatant liquor after the addi- tion of the reagent and before heating with an excess of the acid. When dealing with unknown mixtures it is always best to pro- ceed in this manner. EXPERIMENTS. a. Try above method on a simple salt of Ba. b. Make a mixture of salts of Ca and Ba, recrystallize at once without remov- ing the Ca. From another portion remove the Ca with hot water and recrystal- lize the residue. c. Try a mixture of Sr and Ba. Remove the Sr by treating with HC1 and re- crystallize the residue. d. Try a mixture of Ca, Sr and Ba, recrystallizing at once, then removing in turn the Ca with hot water and the Sr with HC1. e. After having tried the other reactions for Ba fuse some BaSO 4 with K 2 CO 3 and proceed as directed above. 298 ELEMENTARY CHEMICAL MICROSCOPY B. By Means of Oxalic Acid. Read carefully the discussion of this test as given under Calcium and Strontium, pages 292 and 293. Barium oxalate BaC2O4 wH 2 forms large branching aggre- gates, radiating bundles of branching crystallites and sheaves of bristling fibrous needles. Rarely, well-developed monoclinic prisms may be obtained. The branching crystallites are char- acteristic of barium and are never given by calcium or by strontium. Precautions. The solution to be tested should be neutral; even a very little trace of acid is apt to prevent the separation of the character- istic crystals. If no crystals appear after a short time, add a fragment of sodium or ammonium acetate. When calcium or strontium are present the characteristic crystal forms of barium oxalate will rarely be obtained. Re- course may then be had to testing in dilute nitric acid. From nitric acid solutions the barium salt will not separate, while the oxalates of calcium and strontium will slowly crystallize in their usual form. After allowing sufficient time for the complete separation of calcium and strontium, decant, concentrate the solution and add sodium acetate. Barium oxalate now appears, usually in the form of rosettes of thin prisms. Barium oxalate, like the oxalates of calcium and strontium, assumes different crystal forms, according as the test drop is hot or cold. Hot solutions give rise to the production of strongly polarizing orthorhombic plates. Since, in order to facilitate the separation of barium oxalate, sodium acetate has been added, it is well to bear in mind that there is danger of interference from members of the magnesium group. Borates present in the test drop, if in large amount, may prevent the formation of characteristic crystals of barium oxalate. Although chlorides of iron and aluminum have, as has been MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 299 stated, no deleterious influence on the precipitation of the oxa- lates of calcium and strontium, we meet, in the case of barium, with a most interesting and remarkable reaction. Owing to the formation of double oxalates of barium and iron or barium and aluminum, instead of the typical fibrous bundles of needles and crystallites, there are now obtained tufts and bunches of very long exceedingly fine curving hair-like crystals (trichites) of characteristic appearance. The chemical composition and formulas of these compounds have not yet been definitely ascer- tained. In order to obtain this interesting compound, proceed as follows: To the test drop containing barium, add ferric chloride in sufficient amount to impart a faint but distinctly yellow color; then add a fragment or two of sodium or ammonium acetate; stir. The yellow color should now have changed to a reddish tint. Into this drop, thus prepared, cause a drop of oxalic acid to flow. Tufts and sheaves of very fine hairs soon appear. The hairs rapidly grow longer and longer and soon begin to curve in a most peculiar manner. The presence of calcium or strontium, or both, in even large amounts does not appear to have any serious influence on the formation of this double oxalate of barium and iron, save that its separation is often somewhat re- tarded. In such mixtures the oxalates of calcium and strontium first appear in their usual form, then after a time the hair-like tufts of the double oxalate appear. If the quantity of barium is quite small, in proportion to the iron, little rosettes of radiating needles are obtained, separating near the edges of the drop. Aluminum gives rise to the formation of a similar product, but the crystal masses are colorless, while those of the iron salt are light brown. EXPERIMENTS. a. Test a salt of Ba with H 2 C 2 O 4 , in both hot and cold solutions. b. Make a mixture of Ca, Sr, Ba. Add H 2 C 2 O 4 . Repeat the experiment in HN0 3 solution; after a few minutes, decant the clear solution, concentrate slightly and add NaC 2 H 3 O 2 . c. Try the effect of the presence of FeCl 3 on the precipitation of oxalates of Ca, Sr, Ba; first each element separately, then in mixtures of Ca and Ba; Sr and Ba; Ca, Sr and Ba. 300 ELEMENTARY CHEMICAL MICROSCOPY d. If barium borate is at hand, try testing it for Ba. e. Try H 2 C 2 O 4 on a salt of Mg, then add an excess of HC 2 H 3 O 2 to the test drop and examine again. /. Test salts of Zn, Cd, Pb and Ag. BEHAVIOR OF CALCIUM, STRONTIUM AND BARIUM TO OTHER IMPORTANT REAGENTS. The tests already given are generally ample for the proper identification of the alkaline earths, but occasionally problems arise where supplementary or alternate methods are desirable. The following reactions have, therefore, been included both on account of their applicability to the examination of unknown material and because of the further light they throw upon the similarities and differences between the members of the Calcium Group. Behavior with Potassium Ferrocyanide. The reagent is applied by Method 7, page 251, to the test drop acidulated with acetic acid and containing a little ammo- nium chloride. Calcium yields tiny rectangular or square plates. Strontium fails to form a ferrocyanide under the conditions given above. Barium yields large, clear, transparent, yellow rhombs prob- ably belonging either to the orthorhombic or to the triclinic system, depending upon the amount of water of hydra tion. The salts separating are double ferrocyanides to which the following formulas have been ascribed: K 2 CaFe(CN) 6 3 H 2 and K 2 BaFe(CN) 6 5 H 2 O (O?) or K 2 BaFe(CN) 6 3 H 2 (Tr). As usually obtained the barium salt extinguishes parallel to a line drawn through the acute angles of the rhombs. This fact enables the analyst to readily differentiate between the double barium salt and chance separation of the reagent (M). Free mineral acids must be absent. Potassium ferrocyanide, though giving a neat reaction with pure salts of barium, is of little value when dealing with mix- tures. It is then often difficult to avoid the precipitation of MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 301 calcium with the barium, particularly if much ammonium chlor- ide is present, or if much sodium acetate has been added to mitigate the action of mineral acids. From mixtures, strontium may sometimes be precipitated in an amorphous condition if the solution is quite concentrated, and may thus interfere with the test. Pure salts of strontium give, even in very concentrated solutions, only a granular deposit consisting of globular masses, exhibiting no distinguishable crystal form. Magnesium is precipitated from ammoniacal solutions, but neither from acid nor from neutral solutions; hence the pres- ence of this element will not mask the test for barium. In addition to calcium and strontium, there are a number of other elements, which, if present, will either be precipitated in insoluble form or will interfere with the formation of the barium crystals. In this list the most frequently met with will be lead, iron, zinc, rare earths and less often copper, mercury, uranium, and titanium. EXPERIMENTS. a. Crystallize a little of the reagent K 4 Fe(CN) 6 , alone, and determine its optical properties. b. Try reagent on pure salts of Ca, Sr, Ba, using both dilute and concentrated solutions. Try again, this time proceeding as directed above, using HC 2 H 3 O 2 and NH 4 C1. c. Try the reagent on mixtures of Ca and Sr, Ca and Ba, Sr and Ba. d. Try effect of the reagent on salts of Pb, Zn and Fe. Then make mixtures of Ba and these elements and test. e. Make a preparation of K2BaFe(CN)e, measure the angles of the crystals and determine the optical properties of the compound. Behavior with Ammonium or Potassium Bichromate. The reagent is applied to the test drop in solid form, Method 777, page 252. From acetic acid solution, barium chromate BaCr04 is im- mediately precipitated, orthorhombic, in the form of minute light-yellow globular masses, or tiny rods with rounded ends. Strontium chromate will not separate from acid solutions but only from -neutral or slightly alkaline solutions. Calcium is 302 ELEMENTARY CHEMICAL MICROSCOPY precipitated by bichromate from neither acid, neutral nor am- moniacal solutions. The strontium salt of the formula SrCr0 4 appears from ammoniacal solution as exceedingly tiny yellow globulites or dumb-bell-like aggregates; it is dimorphic, being either ortho- rhombic or monoclinic. If the former, it is isomorphous with the barium salt. When this test is used, acidify the dilute drop with acetic acid, then add the fragment of bichromate. Do not stir, and avoid rubbing the glass with rod or wire. Barium chroma te separates at once if present. After several minutes decant if a precipitate has formed. To the decanted solution or clear drop add a small drop of ammonium hydroxide and examine the preparation for dumb-bells of strontium chromate. If both barium and strontium are believed to be present it is best to warm the preparation to cause as complete a precipi- tation of barium chromate as possible before adding the am- monium hydroxide, but care must be taken to avoid unduly concentrating the drop. It is also usually better to allow the ammonium hydroxide to flow into the drop from one side rather than add it directly to the middle of the drop. Normal potassium chromate produces, with barium salts, a precipitate similar to that obtained with dichromate, but is not to be recommended as a reagent because of its property of also precipitating strontium compounds in acid solution. Ordinarily the precipitate of barium chromate is mostly amorphous in appearance. Here and there, however, will be found areas where there are recognizable crystals. A high power is always required for the recognition of the form of the crystals, hence the drop to be studied must be spread out quite thin. Free mineral acids interfere with the test. In addition to barium and strontium, it must be remembered that dichromate will also yield crystalline precipitates with silver, lead, mercury and thallium, but in these cases nitric acid may be present. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 303 EXPERIMENTS. a. Try reaction on salts of Ba, Sr and Ca, in acid, neutral and ammoniacal solu- tions, and both in concentrated and in dilute solutions. b. Try mixtures of Ca and Ba, Sr and Ba; use solutions acidified with HC 2 H 3 O 2 , decant the clear solution, and to it add NH 4 OH. c. Try the reagent upon Ba and Sr salts in HNO 3 solution. Then try it upon Ag, Pb and mercurous salts in HNOs solution. Behavior with Primary Sodium Carbonate. An almost saturated solution of the reagent is added to the dilute ammoniacal test drop by Method /, page 251. Calcium carbonate CaCOa separates in very small disks and rhombs (H or 0). Strontium yields spherulites often of considerable size. Barium separates as minute spider-like aggregates and tiny spherulites, the latter often uniting to form spindles and dumb- bell-like masses. The addition of the reagent in solid form gives nearly as good results. Warming the preparation increases the rapidity of the reac- tion and leads to the formation of better crystals. Unless the test drop is quite dilute an amorphous precipitate results. Ammonium carbonate can be substituted for the sodium salt; the crystals then differ but little if any from those obtained as above, but normal sodium carbonate gives amorphous precipi- tates only and therefore should never be employed. When simple salts of the elements calcium, strontium and ba- rium are employed it is not at all difficult to distinguish between them by testing with primary sodium carbonate (or ammonium carbonate). But if two or more of these elements are present the method fails, characteristic crystals being the exception. In the presence of a great excess of the reagent a double carbonate of calcium and sodium separates, having the formula CaCO 3 - Na^COs 5 H 2 O, which crystallizes in stout monoclinic prisms somewhat resembling the short, thin prisms of calcium sulphate. Strontium and barium prevent the formation of the double salt. 304 ELEMENTARY CHEMICAL MICROSCOPY Elements of the magnesium group interfere. Lithium like- wise interferes. But the chlorides of iron and aluminum and the salts of boric acid have no appreciable effect on the reaction. When in doubt as to the nature of a precipitate formed by the treatment with HNaCO 3 , decant the supernatant solution, which is easily done since the crystals of calcium carbonate adhere to the glass slide, wash the residue, and then add dilute sulphuric acid. If the precipitate is due to calcium, characteristic crys- tals of CaSCX 2 H 2 appear. Primary sodium carbonate is of more value as a group reagent than as an identification test. Moreover, chance formations of crystals of alkali carbonates may be met with in the progress of the systematic analysis of unknown material, particularly when testing for zinc (q.v.). MAGNESIUM. Crystal Forms and Optical Properties of Common Salts of Magnesium. A. ISOTROPIC. B. ANISOTROPIC. Hexagonal. Pyroantimonate. Tetragonal. Fluoride. Orthorhombic. Ammonium-magnesium phosphate; sulphate; primary tartrate. Monodinic. Acetate; chloride; nitrate; primary phosphate ; ammonium-magnesium sul- phate ; potassium-magnesium sulphate ; normal tartrate. Tridinic. DETECTION. A. By Means of Uranyl Acetate and Sodium Acetate. This test has already been described at length under Sodium, Method A, page 275. B. By Means of Secondary Sodium Phosphate in Ammoniacal Solution. For the reaction see Ammonium, page 286. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 305 The detection of magnesium in simple salts is comparatively easy and rapid, since characteristic crystals are readily obtained, but its microchemical identification in complex mixtures is usu- ally a matter of not a little difficulty, in as much as this element is commonly associated with others, closely related, which are prone to interfere with or prevent the formation of typical crystals with the reagents employed for its recognition. Two methods are available, the choice of procedure depending upon the nature of the salts present in the drop to be tested. In all cases where there is a doubt as to the probable composition of the material to be examined, it is best to have recourse at once to the modification II. 1 I. To the solution of the material to be tested, which must not be too concentrated, add several fragments of ammonium chloride; stir; then add a very slight excess of ammonium hy- droxide, and warm the preparation. (If a precipitate results it is best to draw off the clear solution.) To the warm solution add a small crystal of secondary sodium phosphate. Crystals of am- monium magnesium phosphate NH4MgP04 6 H 2 O soon appear. II. To the solution to be tested add a fragment or two of citric acid, stir until dissolved, then add an excess of ammonium hydroxide. Evaporate to dryness. To the residue add dilute ammonium hydroxide. Warm; then add a very small frag- ment of secondary sodium phosphate. Crystals of ammonium magnesium phosphate separate. The crystals of the ammonium magnesium phosphate sepa- rate as skeletons and hemimorphic forms of the orthorhombic system (see Ammonium). It should be remembered that a number of elements are precipitated by phosphates in alkaline solution; the most fre- quently met with in the course of microchemical analyses, either in the substance to be tested, or present as reagents from previous tests, are, doubtless, lithium, members of the calcium and mag- nesium groups, trivalent metals, manganese, nickel, cobalt, tin, lead, silver, copper, and uranium. 2 Of these elements, lithium, 1 Romijn, Zeit. anal. Chem., 37, 300. 2 Most of these elements will generally have been removed in the progress of the analysis before the addition of the sodium phosphate. 306 ELEMENTARY CHEMICAL MICROSCOPY iron, manganese, cobalt and nickel form, with ammonium and phosphoric acid, salts of similar composition to, and isomorphous with, the magnesium salt. The ammonium glucinum phosphate, ammonium zinc phos- phate and ammonium cadmium phosphate are not precipitated in crystal form. The advantage of employing modification II lies in the fact that owing to the presence of ammonium citrate, there is little danger of the interference of the elements listed above. If in following this method, the residue after evaporation is not com- pletely soluble in the ammonium hydroxide solution, it is best, though not essential, to decant the clear liquid before adding to it the sodium phosphate. Reactions I and II work equally well in the cold, but are then a trifle slower. Generally, an amorphous precipitate is at first produced which begins to crystallize in a few seconds. The formation of merely an amorphous precipitate must never be taken as evidence of the presence of magnesium. In the presence of phosphates the detection of magnesium becomes quite difficult, particularly if other elements are present which form phosphates insoluble in ammonium hydroxide. If arsenates are also present, a still further complication arises, for, as we have already seen, double ammonium arsenates of calcium, zinc, etc., are formed, which are isomorphous with ammonium magnesium phosphate. Of course it may happen that in some cases the mere addition of ammonium hydroxide will cause the -separation of character- istic crystals of ammonium magnesium phosphate. Generally, however, it is first necessary to remove the phosphoric acid. This can be accomplished by tin and nitric acid, or by means of ammonium tungstate and nitric acid (see Phosphates, page 380). Precautions. In I, the reaction sometimes fails for lack of sufficient ammo- nium chloride, magnesium hydroxide being precipitated. A slight excess of this salt will do no harm. Both modifications fail if there is an insufficiency of ammo- MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 307 mum hydroxide, for it should be remembered that there must be not only enough ammonium present to unite to form the proper compound, but that this latter salt will not separate save in alkaline solution. It must also be borne in mind that the use of too strong ammonium hydroxide in excess so reduces the solubility of many salts as to cause their separation. Hence it is necessary to avoid, in reactions of this character, deciding too hastily as to the result of a test. EXPERIMENTS. a. Try modification I on a solution of MgSO4, then try it on salts of Fe, Mn r Co, Ni, Al, Zn and Cd. Repeat the experiments, this time adding the HNa 2 PO4 before the NH 4 OH. b. Try modification II upon the same salts and combinations used in a. c. Make mixtures, trying various combinations of the above with members of Groups 1 and II. ZINC. Crystal Forms and Optical Properties of Common Salts of Zinc. A. ISOTROPIC. B. ANISOTROPIC. Hexagonal. Tetragonal. Orthorhombic. Chromate; sulphate. 1 Monoclinic. Acetate; potassium-zinc sulphate. Triclinic. DETECTION. A. By Means of Ammonium Mercuric Sulphocyanate. Apply the reagent by Method /, page 251. This reagent furnishes us with one of the best and the most generally useful methods for detecting the presence of zinc, copper, cadmium and cobalt, and will also furnish evidence of the presence of iron, silver, lead and gold. 1 If formed in the presence of ferrous suipnaie, monociinic. ELEMENTARY CHEMICAL MICROSCOPY For the qualitative examination of simple salts and alloys it leaves little to be desired, but in the analysis of minerals, it is better to employ the carbonate test first, then corroborate with the sulphocyanate reagent. Upon adding a rather concentrated solution of the reagent to a dilute solution of the metals listed above the following results are obtained: Zinc yields an almost instantaneous precipitation of the com- pound Zn(CNS) 2 Hg(CNS) 2 1 in pure white feathery crosses and branching feathery aggregates. These skeleton crystals, when thick, appear black by transmitted light and snow white by reflected light. The normal crystal of the double sulphocyanate of zinc and mercury is said to be a right-angled prism of the orthorhombic system, but under the conditions which obtain in ordinary practice, only skeleton and dendritic forms will be seen. Neither magnesium nor aluminum interfere with this test, save that when magnesium is present in very large amount, the separation of the zinc salt is retarded, and that aluminum under similar conditions renders the skeleton crystals of the zinc salt somewhat less feathery. When zinc alone is present the crystals, as has been stated above, are snow white and of the form described; but if copper is present in minute amount, the crystals of the zinc salt are colored chocolate brown without undergoing any change of form. These brown crystals begin to appear after the white ones have separated. More copper than sufficient to yield the brown tint produces black crystals of modified form; still a greater proportion of copper completely changes the appearance of the crystals, and jet black spheres and botryoidal masses result. Finally a point is reached where crystals of copper mercuric sulphocyanate predominate, accompanied by the black crystals just mentioned. In all cases, however, because of the much lower solubility of the zinc compound than that of the 1 This salt is generally given as anhydrous. Recent work seems to throw some doubt upon this and to indicate the presence of one molecule or less of water of hydra tion. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 309 other complex salts formed, there will always be formed some of the typical uncolored zinc mercury sulphocyanate. Copper alone yields beautiful branching dendrites and radiat- ing masses of acicular prisms, yellowish green in color. The reaction is sensitive and beautiful and constitutes one of the most satisfactory tests available for the identification of copper. The change in color due to the solid solution of the copper salt Cu(CNS) 2 -Hg(CNS) 2 -H 2 O, in the zinc salt is a most inter- esting one and one for which no really satisfactory explanation is yet at hand. The cobalt salt enters into the zinc salt in solid solution to yield light blue crystals. With very small amounts the color is exceedingly faint and the crystal form unchanged, but as the proportion of cobalt increases, the skeleton crystals of the zinc salt become deeper and deeper blue, simpler, less feathery, and gradually assume the color and appearance of the normal cobalt mercuric sulphocyanate. As in the case of the copper-zinc compound, these blue crystals are doubtless cases of solid solu- tion, but the theory of isomorphous mixture is more tenable in this case than in that where copper is present. Cobalt alone yields deep blue-black orthorhombic prisms,. Co(CNS) 2 Hg(CNS) 2 , usually imperfectly developed and unit- ing to form star-like clumps and radiating masses. This con- stitutes a valuable method for differentiating cobalt from nickel, since nickel yields no double sulphocyanate crystals under the conditions which obtain in microchemical testing. Small amounts of zinc in the presence of much cobalt cannot be detected by this reagent. Cadmium yields Cd(CNS) 2 Hg(CNS) 2 in brilliant colorless, probably orthorhombic prisms, usually several times as long as broad but the appearance of these prisms varies with the conditions which obtain at the time of their formation, as, for example, the concentration, depth of the test drop, amount of reagent added, acidity, etc. These variations are, however, not of a kind to render the test doubtful, long prisms, either singly or in groups being the rule. Even a small amount of cadmium destroys the feathery and 310 ELEMENTARY CHEMICAL MICROSCOPY branched character of the skeletons of the zinc-mercury sulpho- cyanate, owing to the formation of mixed crystals, and there generally results crystallites of the shape of an arrowhead. Small amounts of zinc in the presence of much cadmium will usually escape detection. Much nickel modifies the crystals of the double zinc salt in the same manner as cadmium. With much nickel and very little zinc only spherulites are obtained. The presence of both copper and cobalt in a solution contain- ing zinc gives rise to the formation of mixed crystals of very peculiar color and form. These peculiarities are accentuated when cadmium is also present. The experienced worker thus will have little difficulty in detecting a number of elements in one single operation. Manganous salts in excessively concentrated solutions con- taining a trace of free sulphuric acid yield crystals closely re- sembling those of the cadmium double salt. Ferrous compounds, if only in very small amount, do not interfere with the formation of the typical crystals of the zinc salt but in high per cent there will usually be obtained radiating groups or feathery dendrites closely resembling the copper salt. Ferric salts always yield a pink or red color and have no effect upon the zinc compound until a concentration is reached such that a deep blood red color appears. Under such conditions the zinc-mercury sulphocyanate first separates as a deep reddish brown salt, jet black by transmitted light, yet still retaining the typical feathery drendritic form, but in a few seconds these undergo a sudden and remarkable change into masses of curving branching filiform crystals. This is especially marked in test drops containing sodium or ammonium acetate. Lead, unless present in large amount, usually seems to have little or no effect on the zinc reaction. Under some conditions it seems to interfere, however, and it is, therefore, always best to first remove the lead by means of dilute sulphuric acid. Add the acid, decant or filter; evaporate the clear solution to dry- ness; fume off the free sulphuric acid; dissolve in water; add ammonium acetate, and test as above. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 311 Silver gives with the reagent a white amorphous precipitate, soon crystallizing in the form of small, thin, slender prisms with square or oblique ends, somewhat resembling those of the cadmium-mercury salt, but very much smaller than the latter. In the presence of silver the test for zinc is sometimes masked. In such an event, first remove the silver with hydrochloric acid, and test, after evaporation, in the usual manner. EXPERIMENTS. a. Apply the reagent, in the manner indicated, to solutions of pure Zn salts of different degrees of concentration. b. Try in turn pure salts of Cd, Cu, Co, Ni, Ag and Pb. c. To a Zn solution add a very little Cd and test. Repeat the experiment, using more Cd. d. In like manner try mixtures of Zn and Cu; Zn and Co; Zn and Ni; Zn and Fe; Zn and Mg; Zn and Al; Zn and Pb; Zn and Ag. e. Then try more complex mixtures, as, for example: Zn, Cd and Cu; Zn, Cd and Co; Zn, Cu and Co; etc. In each case prepare several slides under different conditions and note well the changes in the appearance in the crystals which separate. B. By Means of Primary Sodium Carbonate. Apply a large drop of a saturated solution of the reagent by Method 7, page 251, to a neutral or very slightly acid drop of the material to be tested. An amorphous precipitate of what is doubtless a basic car- bonate of zinc is usually at first formed and may persist unless the reagent is in large excess; in the latter case, after a few min- utes, a double carbonate of zinc and sodium separates at the periphery of the drop. The crystals of this salt are constant and peculiar to zinc. No other element yields compounds of like appearance. The salt has the formula 3 Na 2 CO 3 '8 ZnCOs'8 H 2 O (Deville). It takes the form of tiny colorless triangles and tet- raheda or three-pointed or five-pointed agglomerates or rarely short stout prisms with pointed ends. The characteristic form upon which to base a decision are the triangles or tetrahedra. The crystals cling tenaciously to the glass, rendering decantation easy. After the removal of the mother liquor the double car- bonate can be dissolved in acid and subjected to other tests. 312 ELEMENTARY CHEMICAL MICROSCOPY It is unfortunate that this, which is one of the most character- istic as well as delicate of the microchemical tests for zinc, should be open to many difficulties. The chief of these lies in the fact that many elements are precipitated as carbonates, and that these often bulky precipitates interfere with or mask the zinc reaction. Among the interfering elements, those most frequently met with are doubtless calcium, strontium, barium, magnesium, cadmium, lead, iron, manganese, cobalt and nickel. Of this list, calcium, strontium, barium and lead will probably have been re- moved by previous treatment with sulphuric acid. Zinc may be separated from the remaining elements of this list by treating with ammonium hydroxide and hydrogen peroxide and finally extracting with a drop or two of moderately concentrated sodium hydroxide solution. To this clear extract primary sodium car- bonate is added. Schoorl has pointed out that the best results are to be obtained from acetic acid solutions of zinc to which normal sodium car- bonate is added. This method is unquestionably the best in the analysis of complex mixtures and when the per cent of zinc present is low. The Behrens method of direct addition of pri- mary carbonate is restricted to simple salts of zinc or to mix- tures known to contain no interfering elements. If only a very small amount of cadmium is present, it is pre- cipitated before the zinc, and by avoiding the addition of an excess of the reagent, decanting the clear liquid and adding to the decanted liquid a fresh portion of the reagent in sufficient quantity, the zinc can be precipitated as the double carbonate. When considerable cadmium is present this method is not feasible. In such an event recourse may be had to ammoniacal solutions, as suggested by Behrens. The test drop is made strongly ammonia- cal and to it primary sodium carbonate is added. Cadmium is immediately precipitated, while the zinc remains in solution. The clear solution is decanted at once. After a few seconds zinc separates from the decanted solution as the double car- bonate in the forms described above. Some little skill and experience is generally necessary in order to obtain good results. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 313 Precautions. Salts of ammonium must be absent or present only in small amounts. The separation of typical crystals is always slow and cannot safely be hastened. It is essential that an excess of the reagent be employed. Failure not infrequently results from a neglect of this precaution. This is particularly true if the test drop is acid. Because of the necessity of adding large amounts of primary sodium carbonate, the test drop must be of greater volume than is usual in micro- chemical testing. EXPERIMENTS. a. Try precipitating Zn in acid, neutral and ammoniacal solutions. b. Test mixtures of Zn and Cd, first in neutral, and then in ammoniacal solu- tions. c. Experiment with Zn in the presence of the interfering elements noted above. C. By Means of Oxalic Acid. The reagent is applied by Method 7, page 251; see Cal- cium, Method B, page 291, Strontium, Method B, Barium, Method B, pages 295 and 298. Zinc yields ZnC 2 O 4 2 H 2 O as small double spherulites, as pseudo-octahedra singly or united in twos, and as thin rhombs. The great majority of the crystals separating usually have their angles rounded. It is rare that a preparation is obtained giving clear-cut crystals. These crystals, when examined with a low power, often bear a striking resemblance to the oxalates of calcium and strontium; therefore to avoid error the alkaline earths should first be re- moved. Cadmium gives clear colorless monoclinic prisms and tabular crystals of the formula CdC 2 O 4 3 H 2 O. The prisms are usually very long and show a marked tendency to form large X's, and radiating aggregates. From concentrated solutions octahedral crystals are also obtained. The typical prisms of cadmium oxa- 314 ELEMENTARY CHEMICAL MICROSCOPY late are seen only when working with comparatively pure salts. In the presence of cadmium the oxalic acid test for zinc is un- reliable. Magnesium salts must be absent, for under certain conditions a double magnesium-zinc oxalate in hexagons and more or less irregular plates will separate. From a number of other precipitated oxalates, zinc oxalate may be separated by dissolving it in ammonium hydroxide and decanting from the insoluble precipitate. Upon evaporation the ammoniacal solution will deposit zinc oxalate, but no longer in the typical form described above, but as masses of radiating curving needles. Unfortunately this method is not applicable in the presence of magnesium and cadmium. Precautions. The solution to be tested should be neutral or only slightly acid, and rather concentrated with respect to zinc. Lead, silver, copper, cobalt, nickel, iron, aluminum, manganese and chromium interfere with the detection of zinc by means of oxalic acid. They should first be removed if reliable results are to be obtained. As stated above, zinc oxalate may be confused with the oxa- lates of calcium and strontium, while magnesium and barium seriously modify its characteristic appearance. EXPERIMENTS. a. Test a pure salt of Zn in dilute and in concentrated solution. Repeat the experiments, substituting Cd for the Zn. b. Make a preparation of ZnC 2 O 4 2 H 2 O; draw off the supernatant liquid, add NH4OH; warm gently and study the preparation. Prepare slides of different degrees of concentration. c. Recrystallize CdC 2 O 4 3 H 2 O in the same manner as the Zn salt. d. Test mixtures of Zn and Cd. e. Recrystallize the mixed oxalates from NH 4 OH. /. Make mixtures of Zn and the interfering elements listed above. Treat the precipitated oxalates with NH 4 OH. Then try Cd in the same manner. g. Try precipitating Zn with HKC 2 O 4 ; K 2 C 2 O 4 ; (NH 4 ) 2 C 2 O 4 . Then try Cd in like manner. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 315 D. By Means of Sodium Nitroprusside. 1 Apply the reagent by Method /, page 251, to a neutral or slightly acid solution. Zinc yields a nitroprusside of low solubility in the form of spherical grains, botryoidal masses or tiny circular disks of a very faint brownish color. Upon standing, a large number of distinct faces develop upon the spheres (combination of cube and dodeca- hedron ?). These crystals are isotropic. The formula of the compound has not yet been established; that of the reagent can be written Naa NO Fe (CN) 5 - 2 H 2 O. If the zinc merely re- places the sodium, we should obtain Zn - NO Fe (CN) 5 xH 2 O, or, on the other hand, we may be dealing with a sodium-zinc nitro- prusside. In the presence of free mineral acids there is a ten- dency for zinc nitroprusside to separate in tiny squares and stout prisms or in fusiform rods. A moderate amount of free mineral acid does not appear to prevent the reaction but retards the appearance of the crystals. Much acetic acid (or acetates) retards the separation even more. Heat hastens the reaction, but warming does not appear to be of value in obtaining a better development of the crystal form. Cadmium yields tiny rough globulites, octahedra with rough, corrugated or even bristling faces, and drusy masses. Cadmium nitroprusside polarizes strongly and the largest of the crystals exhibit brilliant polarization colors. Mixtures of zinc and cadmium yield rough globulites, most of them anisotropic. Manganous salts give globulites similar in all respects to those obtained with zinc; they appear later and rarely develop to as large a size or exhibit the many faces. Like the zinc salt they are isotropic. In ordinary routine analysis it is practically im- possible to distinguish between zinc and manganese. Copper yields an immediate amorphous pale blue precipitate. Often this shows a tendency toward the formation of star-like skeleton crystals. Mixtures of copper and zinc yield, in addition to an amorphous precipitate, the spherical grains of the zinc salt, but in this case there is a tendency toward spherulites, tiny. 1 Bradley, Am. J. Sci., 22 (1906), 326. 316 ELEMENTARY CHEMICAL MICROSCOPY bristling masses and tiny crosses and stars, closely resembling the forms obtained with cadmium. They differ, however, from the cadmium salt in that they do not polarize. Nickel gives a light green amorphous precipitate; cobalt a similar pink one; while iron, if heated, yields a yellow deposit. Mercurous salts (nitrate) give a gelatinous amorphous mass of a yellowish tint. Mercuric salts and those of silver, lead, tiri, antimony, bismuth, aluminum, magnesium and the alkaline earths appear to give no precipitates and to yield no crystals even in concentrated solution or upon evaporation. Precautions. The solution should be neutral or but faintly acid and should be moderately concentrated with respect to zinc. If no result is obtained upon the first test, make a second, employing a considerably greater amount of. the unknown substance. Heating the preparation hastens the reaction. If a precipitate is obtained, zinc, cadmium, copper, nickel, cobalt, iron or manganese are present and, conversely, if no pre- cipitate appears, these elements must be absent. Sodium nitroprusside is thus a convenient group reagent. EXPERIMENTS. a. Try the reagent upon several different concentrations of Zn. b. Try with Cd, then with mixtures of Zn and Cd. c. Try salts of Cu, Ni, Co, Mn, first as pure salts, then as mixtures with Zn. CADMIUM. Crystal Forms and Optical Properties of Common Salts of Cadmium. A. ISOTROPIC. B. ANISOTROPIC. Hexagonal. Iodide, ammonium-cadmium bro- mide; ammonium-cadmium chloride; potas- sium-cadmium chloride. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 317 Tetragonal. Orthorhombic . Bromide. Monoclinic. Acetate; chloride; sulphate. Tridinic. DETECTION. A. By Means of Ammonium Mercuric Sulphocyanate. Read Method A, Zinc, page 307. The prismatic crystals of Cd (CNS) 2 Hg(CNS) 2 are, in a similar manner to the zinc salt, colored a faint chocolate brown by traces of copper. This brown color intensifies with an in- crease in the amount of copper. When considerable copper is present, the copper double salt first separates, since it is slightly less soluble than the cadmium compound; then mixed crystals form, in which the copper apparently predominates over the cadmium. These mixed crystals are of a deep bluish green color. By this time most of the copper and but little of the cadmium have been precipitated, and the concentration has also reached such a point that the cadmium double salt begins to separate in the crystal forms described on page 309. These are, however, still mixed crystals, for they are colored brown by the small amount of copper still in solution. As in the case of the zinc reaction, iron may sometimes color the cadmium salt a reddish brown. Cobalt colors the cadmium salt blue. Much cobalt gives an intense blue color and alters the crystal form. Magnesium and aluminum have even less effect than in the case of zinc. Before testing for cadmium with the sulphocyanate reagent, it is best to first remove any lead or silver which may be present. If a small amount of zinc is also present, mixed crystals con- taining zinc and cadmium first separate, whose crystal form can be described as non-feathery skeletons; soon after this the cadmium double salt separates in its typical form. In order that this sequence shall be brought about, it is best to employ a solution somewhat more dilute than when zinc is known to be absent. Much zinc usually prevents the formation of any of the 318 ELEMENTARY CHEMICAL MICROSCOPY prismatic crystals of the cadmium salt, only mixed crystals resulting. Precautions. Cadmium salts of the organic acids, as, for example, cadmium acetate, fail to yield a satisfactory test. It is therefore best to evaporate the unknown with nitric acid and drive off the excess of acid before adding the sulphocyanate reagent. It follows that the addition of sodium or ammonium acetate to very acid solu- tions to lessen the effect of the mineral acid is in this case unwise. It is better to evaporate to dryness. B. By means of Oxalic Acid. Read Method C, Zinc, page 313. The typical crystals of cadmium oxalate CdC204 3 H^O con- sist of long, clear, colorless, monoclinic prisms, singly, in X's, or in clusters. The obliquely truncated ends constitute a dis- tinctive feature. Manganous oxalate MnC2O4 3 H^O separates in groups of radiating prisms, which the careless observer sometimes con- fuses with the cadmium salt or vice versa. The ends of the prisms of the two salts are quite different however in appearance. C. By Means of Sodium Nitroprusside. See Zinc, Method D, page 315. MERCURY. Crystal Forms and Optical Properties of Common Salts of Mercury. A. ISOTROPIC. B. ANISOTROPIC. Hexagonal. Tetragonal. Mercurous bromide, chloride and iodide; mercuric cyanide; red mercuric iodide. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 319 Orthorhombic. Mercuric bromide ; mercuric chlo- ride; yellow mercuric iodide. Monoclinic. Mercurous and mercuric nitrates. Tridinic. DETECTION. A. As Metallic Mercury by Sublimation. Heat upon a piece of platinum foil or upon a glass slide a little anhydrous sodium carbonate until all the moisture it con- tains has been expelled, cool, powder and mix a very small amount with a little of the material to be examined transfer to a small tube of hard glass not over 2 millimeters in internal diameter, thin-walled and sealed at one end. Jar the mixture down so as to obtain clean walls. Heat gently over the flame of a Bunsen burner turned down to a flame i centimeter high. The mercury compound will be decomposed and tiny globules of metallic mercury will condense upon the walls of the tube. Examine under the microscope. With a stiff hair or glass rod drawn down to a hair gently rub the ring of sublimate. Examine again. The mercury will have united into larger globules. Introduce into the tube two or three small fragments of iodine. Then insert the open end of the tube into a piece of cork; warm the iodine very gently and set the tube aside for a few minutes. Yellow and red mercuric iodide will be formed. Warming again will hasten the reaction and cause the sublimation of some of the mercuric iodide. Rectangular and rhombic plates and dendritic masses of both the vermilion colored iodide and the yellow modification will be obtained. No other known element gives a reaction even remotely re- sembling this one. From large volumes of liquid the mercury may be removed by acidifying with hydrochloric acid and dropping in a steel needle around which has been wound a tiny spiral of thin gold foil. The deposited mercury amalgamates with the gold. The electrolytic couple is lifted out after some time, washed, the gold foil removed, dried, placed in a subliming tube and the mercury expelled by heating. The sublimate is then characterized as above. 320 ELEMENTARY CHEMICAL MICROSCOPY From drops containing moderate amounts of mercury, the metal may be separated by a fragment of magnesium, or it may be deposited upon a bit of copper. If in the latter case the spot of deposit be rubbed it becomes silvery white. If the coated copper is placed in a subliming tube and heated the mercury will be volatilized and will condense in characteristic globules. EXPERIMENTS. a. Test several mercurous and mercuric salts by heating them with Na 2 CO 3 . Examine the sublimates. Rub them gently with a hair-like glass rod and note that the globules unite. b. Obtain a deposit of Hg upon a tiny bit of Cu foil i millimeter by 3 milli- meters by heating in a drop of a solution of an Hg salt acidified with HC1. Dry and sublime. c. Introduce a fragment of iodine in one or more of the tubes, warm gently and allow to stand about five minutes. Examine for crystals of Hgl2. B. Differentiating between Mercurous and Mercuric Salts. Add Hydrochloric Acid. With mercuric salts there is no precipitation. Mercurous salts give an immediate amorphous precipitate of a white chloride HgCl. Under unusual conditions and exceedingly dilute solutions, mercurous chloride may some- times be obtained in the form of slender needles. To charac- terize the white precipitate, draw off the supernatant solution and add to the residue a drop of dilute ammonium hydroxide. A black compound of the formula NH 2 Hg 2 Cl is immediately formed. Examined with a J inch or an 8 millimeter objective the black compound is seen to consist of a mass of tiny acicular crystals, tiny squares, crosses and fusiform grains. EXPERIMENTS. a. Precipitate HgCl, examine with the microscope. b. Add NH 4 OH to the white precipitate and examine again. Add Potassium Bichromate and Nitric Acid. To the drop to be tested add nitric acid. Place nearby, a drop of solution of bichromate. Warm the drops over the micro-flame and while MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 321 hot cause the bichromate to flow into the test drop. Mercurous salts yield characteristic crystals. Mercuric salts do not. 1 There are generally formed with mercurous salts a number of different compounds. There first separates a dark red granular precipitate, soon changing into dark red crosses, bundles of irregu- lar crystals and peculiar dendrites and skeleton masses. Later yellow crystallites appear. In any given test the appearance of the precipitate both as to crystal form and color will depend upon the concentration of the drops, the degree of acidity and the temperature. Mercuric salts give no such precipitates and no crystalline compounds will appear unless the preparation is allowed to evaporate practically to dryness. There will then appear light yellow feathery dendritic and radiating branching moss-like masses. Lead yields slender yellow monoclinic prisms, seldom grouped in masses. This element unless present in excess does not appear to seriously interfere with the test for mercury. Silver separates in dark red pleochroic plates and scales which may often mask the mercury compounds. EXPERIMENTS. Test as above both mercurous and mercuric salts with and without HNOs present in both cold and hot solutions. C. Add to a Drop of the Material a Tiny Fragment of Potas- sium Iodide. See Method ///, page 252. Mercuric salts yield vermilion colored mercuric iodide; mercurous salts a heavy bright yellow amorphous precipitate somewhat resembling lead iodide in color but instead of being in plates always agglutinated in a formless mass. With mercuric salts we obtain one of the best and most satis- 1 Bichromate added to hot unacidified HgCl2 solutions causes the separation on cooling of hard star-like masses of crystals. According to Millon (Ann. chim. phys. (3) 18, 388) this compound has the formula HgCl 2 K 2 Cr 2 O 7 . Ammonium bichromate gives orthorhombic six-sided prisms of the compound HgCl 2 '3(NH4) Cr 2 O 7 . 322 ELEMENTARY CHEMICAL MICROSCOPY factory tests for mercury. At the moment the potassium iodide strikes the drop a white or pinkish cloud appears, rapidly chang- ing to yellow then to brilliant red. The mercuric iodide HgI 2 first formed is very soluble in excess of the reagent forming the soluble compound HgI 2 2 KI. The precipitate therefore appears as an ever-widening circle about the fragment of solid reagent until the latter is completely dissolved. If the outer edge of the brilliant red circle is now examined with a moderately high power it will be seen to consist of tiny ruby red rhombs and rods to- gether with more or less spherical masses and imperfect rosettes. Precautions must be taken to avoid adding an excess of reagent; otherwise no permanent separation will take place. In order to avoid the possibility of error it is always well to add a fragment of copper sulphate, which will take up the excess of iodide and cause the separation of the mercuric salt. EXPERIMENTS. See under Lead, Method A, page 325. D. Mercuric Salts can be detected through the Formation of Double Sulphocyanates. This test is the reverse of that employed for the detection of Zinc (Method A , page 307) ; of Copper (Method A , page 339) ; or of Cobalt (Method A, page 366), to which the student is re- ferred for details. Add to a small test drop (which must not contain much free mineral acid) a fragment of ammonium sulphocyanate about the size of a pinhead. Stir until dissolved. Place next this drop a tiny drop of water in which is dissolved a very little zinc sul- phate. Cause the test drop to flow into the zinc solution. Char- acteristic crystals of zinc-mercury sulphocyanate will appear. Instead of zinc sulphate, copper sulphate or cobalt nitrate may be employed. With simple mixtures, this test is a very beautiful one, but with complex material it is sometimes difficult to adjust the con- ditions especially as regards the quantity of ammonium sulpho- cyanate required. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 323 EXPERIMENTS. a. Test as above HgCl 2 , using ZnSO 4 . b. Try again, this time introducing a trace of CuSO 4 . c. Try this test with CuSO 4 but with no ZnSO 4 present (which method is most satisfactory?). LEAD. 1 Crystal Forms and Optical Properties of Common Salts of Lead. A. ISOTROPIC. Nitrate (I). B. ANISOTROPIC. Hexagonal. Iodide. Tetragonal. Orthorhombic. Bromide; chloride; 2 sulphate; tar- 1 f crate. Monoclinic. Acetate; chromate; sulphocyanate, Triclinic. DETECTION. A. By Means of Potassium Iodide. Apply the reagent, by Method ///, page 252, to the test drop slightly acidified with nitric acid. Lead iodide PbI 2 is at once formed as a bright yellow precipi- tate in a circular band about the reagent fragment. The circle gradually becomes larger and larger and at its outside circum- ference beautiful hexagonal plates appear. These plates and flakes of lead iodide appear greenish or brownish yellow by transmitted light, sometimes even gray, according to their thick- ness. By reflected light lead iodide plates glow and glisten and display the iridescent colors of thin films, an extremely charac- teristic feature of this salt. These hexagons of lead iodide do not belong, according to 1 Lead, silver and copper are introduced at this point rather than in their proper position in the Periodic System because of their close relations in qualita- tive analysis. 2 Recrystallized from hot water PbCU is pseudohexagonal. 324 ELEMENTARY CHEMICAL MICROSCOPY Behrens, to the hexagonal system, as usually stated, but are probably only pseudohexagonal and in reality orthorhombic. From neutral solutions containing lead in the form of lead acetate, potassium iodide will generally precipitate, in addition to the normal iodide, basic iodides of variable composition, such asPbI 2 .PbO; PbI 2 . 2 PbO (?). Lead iodide can be recrystallized from hot water, best if acidi- fied with nitric acid. On cooling, large, beautifully formed hexagons separate. A large drop of water is necessary in order that good results may be obtained. Heated with hydrochloric acid lead ioolide dissolves, and on cooling crystals of the normal iodide Pbl2, the normal chloride PbCl 2 and a chloriodide PbCl 2 - PbI 2 or 2 PbCl 2 - PbI 2 (or both) will separate. The chloriodides appear in the form of needles of a faint yellow color. Silver iodide separates as a yellowish amorphous mass insoluble in hot water and in hot nitric acid. Mercuric iodide takes the form of red rhombs. Mercurous salts acidified with nitric acid usually give in addition to the heavy precipitate of mercurous iodide the ruby colored rhombs of the mercuric salt. If cuprous salts are present a white granular precipitate of cuprous iodide is formed and iodine is set free. Cupric salts will behave similarly. Thallium is precipitated as an exceedingly fine granular pre- cipitate. Antimony and bismuth salts interfere with the reaction for lead. These elements yield with potassium iodide, double iodides which separate in neat, well-formed crystals. Solutions containing lead, antimony and bismuth, when treated with potassium iodide, yield a dark reddish brown, sandy precipitate wholly unlike in appearance anything obtained with the different elements alone. Boiling the mixed product with water will generally cause a partial decomposition, and on cooling hexagons and irregular plates of lead iodide will appear. In the presence of a little bismuth, lead iodide separates as orange red disks and plates, or the iodide scales may even appear crimson in color. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 325 Precautions. An excess of the reagent must be avoided, otherwise the pre- cipitate at first formed will be dissolved because of the formation of a double iodide of the composition PbI 2 2 KI o;H 2 O. 1 Not infrequently colorless crystals of this double iodide will be seen in the immediate neighborhood of the reagent fragment. The addition of a drop of water will usually cause the decomposition of the double salt and a precipitation of the normal iodide. Double iodides of lead with many elements are known, most of them crystallizing readily, 2 but it is not often that there will be a sufficient separation of these interesting salts to interfere in any way with the detection of lead. Too much nitric acid in the water employed for recrystallizing the precipitate of lead iodide will cause partial decomposition and consequently the separation of colorless octahedra of lead nitrate. EXPERIMENTS. a. To a test drop containing Pb(NO 3 )2 add KI. Study the preparation, then add a drop of water and heat to boiling. After the drop has cooled, study it again. Repeat the experiment, but this time use an excess of KI. Try again in acidified solutions. b. In like manner test a preparation of Pb(C2H 3 O2)2. c. Make a preparation of PbSO*. Decant the mother liquor, add to the sul- phate residue a drop of water, acidify with HNO 3 , then add a fragment of KI. After a few seconds examine the preparation. d. Make a mixture of Pb and Ag, test with KI. Then try in turn mixtures of Pb and Sb; Pb and Bi; Pb, Sb and Bi; Pb and Cu; Pb and Sn. e. Test a preparation of HgCl 2 . Then one of Hg(NO 3 ) 2 . Make a mixture of Pb(N0 3 ) 2 and Hg(N0 3 ) 2 and test. B. By Means of Hydrochloric Acid. Apply the reagent by Method ///, page 252, to the test drop acidulated with a little nitric acid. This method of adding the reagent is not so good as allowing two drops to flow together but is adopted so as to conform to that for testing for silver and mercury. 1 Brooke, Ch. N., 1898, 191. 2 See Mosnier, Ann. chim. phys. (7) 12, 374; Comptes rend., 120, 444. 326 ELEMENTARY CHEMICAL MICROSCOPY Lead chloride PbC^ separates at once in the form of char- acteristic, white, long acicular crystallites belonging to the ortho- rhombic system. There are also seen feathery dendritic X's and long irregular ragged prisms. The appearance of the lead chloride separating varies with the concentration of the solution being tested and with the nature of the substances present. If the test drop is not sufficiently concentrated the lead chloride will not separate at once in the form of the characteristic crystallites, but will appear more slowly, prismatic forms being the rule. This question of con- centration becomes a most important one if the substance con- tains salts with which lead chloride can unite to form double salts, as for example chlorides of the alkali metals and ammonium, for in such an event dilute or even moderately concentrated drops fail to yield recognizable forms. Indeed it may be said that testing for lead with hydrochloric acid is not advisable in the presence of members of Groups I and II. In neutral solutions of lead acetate there may be precipi- tated in the presence of members of Group I and no excess of the reagent, colorless, highly refractive prisms of the formula Pb(OH)Cl (n = 2.08 to 2.16) belonging to the orthorhombic sys- tem but some tunes also appearing as monoclinic prisms. Lead chloride is slightly more soluble in water containing a little nitric acid than in pure water, hence the separation of lead as chloride is never complete. Lead chloride differs from the chlorides of silver and mercurous mercury in being easily soluble in hot water, thus affording a simple method of separation. On cooling, the lead chloride no longer appears in the forms stated above but assumes that of thin pseudohexagonal prisms, rhombs and hexagons. Recrystallized in the presence of Group I, double chlorides result, which generally separate more slowly. The crystal form is quite different from that of the normal salt. It is quite impor- tant that the student should be familiar with at least the double chloride of cesium and lead (cesium chloroplumbate) , since this compound not infrequently makes its appearance when testing for tin with cesium chloride and is quite apt to puzzle the beginner. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 327 Alkalies convert lead chloride into a basic chloride to which the formula PbCl 2 3 PbO 4 H 2 O is generally assigned. Thallous salts yield with hydrochloric acid star- and cross-like crystallites differing considerably from those given by lead. There is little danger of confusing these two elements, since re- crystallizing thallous chloride from hot water, in which it, like lead chloride, is soluble, yields well-formed cubes. In the presence of chlorides of antimony and bismuth complex chlorides of low solubility are sometimes formed, against which the analyst should be on his guard. Silver gives an amorphous precipitate and mercurous salts a fine granular one without resolvable structure. EXPERIMENTS. a. To a drop of a concentrated solution of Pb(NOa)2 add a drop of dilute HC1 in the manner described above. Make several other preparations varying the concentration of the test drops. b. Recrystallize a preparation of PbCl2 by heating to boiling with a large drop of water. c. Recrystallize a preparation of PbCl 2 in the presence of NaCl, another in the presence of KC1; of NRtCl; of CsCl. d. Test a solution of Pb and Sb. Then one of Pb and Bi. Then one contain- ing all three elements. e. To a preparation of PbCl 2 add a drop of NH 4 OH. C. Through the Formation of a Triple Nitrite of Lead, Cop- per and Potassium. To the moderately concentrated neutral test drop add a trace of acetic acid, then a fragment or two of sodium acetate and of copper acetate. Stir. Then add a fragment of potas- sium nitrite. There is formed the salt K 2 CuPb(NO 2 )6 as tiny squares or rectangular plates, or tiny cubes and rectangular prisms which are brown by reflected light, jet black by transmitted light. The crystals appear to be isometric. In this salt the potassium may be replaced by rubidium, yield- ing a compound of lower solubility, or by cesium which will give a salt of less and finally by thallium, one of least solubility and therefore the test of highest delicacy. These salts are probably 328 ELEMENTARY CHEMICAL MICROSCOPY isomorphous. The size of the crystals obtained decreases as their solubility decreases. This test is a most convenient one if alloys or substances sus- pected of containing both lead and copper are being examined. It is then only necessary to add to the solution, sodium acetate, potassium nitrite and acetic acid. If, after waiting a reasonable time, no triple nitrite separates, cesium chloride or thallous nitrate can be added. The nickel salt also forms squares, rectangles and cubes but these are light brown by transmitted light not black. Cobalt is immediately precipitated by potassium nitrite as a very insoluble double nitrite of potassium and cobalt in the form of a reddish brown powder, or in well-defined very tiny cubes and octahedra. The triple nitrite may be written thus: 2 KN0 2 Cu(NO 2 ) 2 Pb(N0 2 ) 2 . Precautions. In very dilute solutions the test fails unless rubidium or cesium chlorides are added because of the too great solubility of the potassium salt. Concentration may sometimes yield the typical black crystals. The addition of an excessive amount of potassium nitrite is objectionable because of the fact that the triple nitrite is quite soluble in solutions of this reagent. On the other hand, it is essential that the amount added be very slightly in excess of that called for by theory. It is therefore necessary to proceed somewhat cautiously. Add a tiny fragment of nitrite, then after waiting a few moments, if no crystals appear add a little more. Too concentrated solutions of lead yield sandy black precipi- tates requiring recrystallization. Recrystallization can be effected by adding to the preparation a little water, a trace of acetic acid and a slight excess of potassium nitrite, then heating the preparation to boiling. Good crystals should appear on cooling. Free mineral acids must be absent. When the amount of lead is relatively great and cesium chloride MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 329 is added to increase the delicacy of the reaction a double chloride of cesium and lead is formed which separates simultaneously with or even before the triple nitrite. EXPERIMENTS. a. Test a preparation containing Pb. b. Try another preparation, this time introducing RbCl. c. Try again with CsCl. d. By a series of careful dilutions determine the limit of the precipitation of Pb as the K salt, the Rb salt and the Cs salt. e. Test a mixture of Pb and Ni; Pb and Co; Pb and Ag. D. By Means of Metallic Zinc. Apply the fragment of metal to the center of the drop to be tested; see Method ///, page 252. The characteristic appearance of the different metals when separated from their solution by an element higher in the electro- chemical series is often quite sufficient to enable the analyst to identify it. The student is already familiar with these peculiari- ties through the experiments performed as outlined on page 254. Lead yields beautiful long stiff many branching more or less fern-like dendrites, whose side arms are usually at right angles to the main stem or rib. Only portions of the formation show bright metallic reflections. The chief characteristic of the "lead tree" is a long fairly straight trunk or rib with side dendrites of irregular length. Of " trees" formed by other metals that of silver most nearly resembles that of lead, but is more delicate, more branching, with side formations at angles other than 90 degrees and exhibits splendid silvery white metallic reflections. Tin somewhat resembles silver but the side arms of the " trees" are very oblique and parallel one with another, that is, the paral- lelism extends across the main axis or rib. The reduction is slower with tin. None of the remaining metals yield long loose fern-like or tree- like forms. Bismuth gives black and gray feathery and mossy dendrites with sharp-pointed ends with a characteristic curving tendency of the ends of the clusters. The mossy dendrites appear 330 ELEMENTARY CHEMICAL MICROSCOPY jet black by transmitted light, grayish by reflection, their growth is rapid and vigorous, finally occupying the entire area of the drop, and is characteristic of bismuth. Antimony yields black mossy dendrites but rarely feathery or curving; they appear more granular in structure. Copper separates as black, compact stout mossy masses with somewhat tabular or angular ends. Cobalt resembles copper somewhat but forms dendrites less readily. Nickel can be made to yield a crystalline deposit only with great difficulty; only small mossy patches are usually obtainable. Gold yields very compact mossy or granular dendrites and irregular botryoidal black masses which soon exhibit the char- acteristic golden yellow reflections of the metal. Precautions. To obtain the best results, the solutions should be practically neutral or only very slightly acid, otherwise the rapid evolution of hydrogen will cause the disintegration of the deposited crystal masses. If free mineral acid is present add sodium acetate. Use only cold solutions. Employ only a very minute fragment of zinc, otherwise the area of metal upon which deposition can take place is so great that really characteristic growths will not be obtained. In general a moderate concentration is essential to the forma- tion of satisfactory dendrites. EXPERIMENTS. If a number of elements have not already been tested under Method III, page 252, try a fragment of Zn in drops of solutions of salts of Pb, Bi, Sb, Sn, Cu, Cd, Pt, Au and Hg. SILVER. Crystal Forms and Optical Properties of Common Salts of Silver. A. ISOTROPIC. Chloride (I); bromide (I); iodide (I or H). MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 331 B. ANISOTROPIC. Hexagonal. Iodide; 1 secondary arsenate; sec- ondary phosphate. Tetragonal. Orthorhombic. Chromate; nitrate; nitrite; sul- phate; potassium-silver iodide. Monoclinic. Triclinic. - Bichromate. DETECTION. A. By Means of Hydrochloric Acid. Apply the reagent by Method III A, page 254, to the test drop previously acidulated with nitric acid. If silver is present an immediate precipitate should result. Examine under the microscope. Silver chloride is so insoluble in water that it is thrown down as an amorphous mass. If the precipitate is wholly crystalline, either silver is absent or else present in very small amount. In order to identify silver in an amorphous precipitate it is necessary to recrystallize it. Before so doing it is always advisable, and often necessary, to first remove the solution from the precipitate and wash the latter. If the hydrochloric acid has been carefully added and the drop not stirred, it is easy to draw off the clear solution from the curdy, heavy precipitate of silver chloride. When the amount of pre- cipitate is very small it is best to have recourse to the centrifuge to accomplish the separation. After removing the supernatant liquid, wash the precipitate once or twice with hot water acidified with nitric acid. The washed precipitate is then recrystallized from concentrated hydrochloric acid, or from ammonium hy- droxide. To the precipitate of silver chloride, at the corner of a slide, add a drop or two of concentrated hydrochloric acid, and heat the preparation over the micro-flame. If the precipitate is not completely dissolved, rapidly draw off the hot acid, without exercising any great care. On cooling, tiny crystals of silver chloride separate. Octahedral crystals predominate. 1 Upon heating, Agl becomes isometric. 332 ELEMENTARY CHEMICAL MICROSCOPY To the washed precipitate add one or two drops of strong ammonium hydroxide. After a second or two of contact, draw off the ammoniacal solution from any undissolved precipitate. Do not heat the preparation. Allow the preparation to stand. Almost immediately the drop becomes turbid around the edges, because of the separation of minute crystals of silver chloride; these crystals increase slowly in size, but are always very small, requiring a moderately high power for distinguishing their form. From ammoniacal solutions silver chloride seems to separate almost invariably in the form of cubes and hexagonal and rec- tangular plates. Only rarely are octahedral crystals obtained. Of the two recrystallization methods, that with ammonium hydroxide will be found to be the better, as well as also the more convenient, because of the greater solubility of the precipitate in this reagent, and because the employment of ammonium hydroxide eliminates many interfering substances. Lead chloride is precipitated in the form of white acicular crystals, irregular crystaMites and X-like dendrites, soluble in hot water and therefore easily removed. Mercurous salts yield a granular precipitate, but sometimes minute needles. Recrystallized from concentrated hydrochloric acid tetragonal crystals may be obtained but no cubes and part of the salt is converted into soluble mercuric chloride. Mercu- rous salts therefore interfere with the satisfactory detection of traces of silver by masking the tiny cubes of silver chloride. Thallous salts yield cubes and stars. Treated with ammonium hydroxide, silver chloride dissolves with the formation of the compound AgCl'2NH 3 (Isambert). If mercurous chloride is present the precipitate turns black under the action of the reagent, an insoluble compound being formed which Barfoed has shown to be a mixture of metallic mercury and the compound HgNH 2 'Cl. If, therefore, silver chloride is present only in traces in a precipitate consisting chiefly of mercurous chloride, ammonium hydroxide may dissolve practi- cally no silver chloride, since the finely divided metallic mercury may reduce the greater part of the silver salt to metallic silver. (Silver follows mercury in the electrochemical series.) Under MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 333 such conditions it is necessary to exercise the greatest care in order to avoid missing the little silver which is present. Elements forming oxychlorides may under exceptional con- ditions be precipitated with the silver. It is also well to bear in mind that the addition of hydrochloric acid may force back the dissociation of certain salts to a degree causing the separation of a solid phase. Precautions. When working with concentrated hydrochloric acid or strong ammonia, great care must be used to avoid spoiling the micro- scope and objectives. It is essential to work rapidly. The drop is acidified with nitric acid because the presence of this reagent favors the agglutination of the particles of silver chloride, and hinders at the same time the precipitation of oxy- chlorides, etc. Decanting after precipitation is advisable, since the crystal form of silver chloride is changed by many compounds when the former is crystallized in the presence of the latter. Still other compounds completely ruin the test. Although there is, of course, danger of the occlusion of some of these objectionable salts by the silver chloride, this difficulty is reduced to a mini- mum by avoiding too concentrated test drops and washing the precipitate. Washing the precipitated silver chloride with hot water re- moves the greater part of the lead chloride which may have been carried down with the silver. EXPERIMENTS. a. Precipitate with dilute HC1, a test drop containing AgNO 3 . Separate and wash the precipitate; then recrystallize it by the above described method, using concentrated HCl. Then repeat the experiment, using NHUOH as the solvent. b. Make a mixture of Ag and Pb, test by both recrystallization methods. c. In like manner test a mixture of AgNO 3 and HgNOs. d. Precipitate with HCl a test drop containing Pb and Ag; recrystallize the precipitate without drawing off the solution. In like manner test mixtures of Ag and Zn, Ag and Cd, Ag and Sb, Ag and Pt, Ag and Sn, Ag and Cu. e. Try recrystallization of AgCl in the presence of phosphates, in the presence of sulphates and in the presence of molybdates. 334 ELEMENTARY CHEMICAL MICROSCOPY B. By Means of Ammonium Bichromate. Acidify the test drop with nitric acid. Add a fragment of the reagent at the center. Allow to stand a few seconds. Dark red triclinic pleochroic crystals of the formula Ag 2 Cr 2 7 appear in the form of thin plates, having a rectangular or more or less symmetrical coffin-like outline. Aggregates of irregular broken scales are also abundant. Insufficiently acidified drops or those which are very concen- trated yield as the first crop of crystals, tiny rods or needles so dark colored as to appear black; after a time there will generally separate in addition to these rods, the characteristic plates and scales mentioned above. Cold solutions of lead yield only a bright yellow amorphous precipitate. But from hot solutions, thin but long and slender monoclinic prisms are formed, not however of lead bichromate but having the composition PbCrO 4 . Lead chroma te is soluble in sodium hydroxide solutions. Mercurous salts yield with ammonium bichromate, in solutions acidified with nitric acid, a number of different compounds (see Mercury) varying in composition and appearance according to the conditions which obtain. There is, however, little danger of confusing these salts with the silver bichromate, since they all appear as dark red crosses and bundles of irregular outline. These compounds may, however, seriously interfere with the recognition of silver if the latter is present only in traces. Mer- curous chromate is insoluble in sodium hydroxide, a distinction from lead. 1 Silver bichromate can be recrystallized from hot water, but better results follow the use of dilute nitric acid or of ammonium hydroxide. From hot nitric acid very beautiful preparations can be obtained. According to some investigators the crystals which separate on cooling from a hot neutral aqueous solution of the bichromate precipitate are not silver bichromate, but normal silver chromate Ag 2 CrO4. Ammonium hydroxide dissolves silver bichromate with ease. 1 If, however, only a minute quantity of sodium or potassium hydroxide is used, a red basic chromate of lead results. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 335 The crystals separating from the ammoniacal solution are, accord- ing to some chemists, complex salts, containing one or more mole- cules of NH 3 . The recrystallized product separates in the form of needles, skeleton crystals and masses resembling lichens. Unless the original precipitation was made in nitric acid solu- tion both strontium and barium may, under unusual conditions, be precipitated. It is well to bear this in mind when recrystal- lizing from ammonia. In the presence of much lead the reaction often fails. Instead of the dark red salt, small yellow prisms of entirely different appearance separate. In such an event either first remove the lead with a drop of dilute sulphuric acid and then add the bichro- mate, or else add, immediately after the fragment of the reagent, a drop or two of dilute sulphuric acid. Usually in a short time good crystals can be obtained. The use of sulphuric acid in connection with the bichromate complicates matters, since the crystals separating in the presence of the silver sulphate formed in the reaction may be either those of the salt Ag 2 Cr 2 O7 or the salt Ag 2 Cr04; the latter compound is usually formed when the amount of nitric acid is small and that of silver sulphate large. Normal silver chromate is isomorphous with normal silver sul- phate, normal silver selenate, and anhydrous sodium sulphate; all are to be referred to the orthorhombic system. Because of this isomorphism of the sulphate and chromate very interesting and instructive preparations may be obtained. Silver sulphate separates from solution generally in the form of highly refrac- tive, transparent, colorless, rhombic octahedra, but in the pres- ence of silver chromate these colorless octahedra increase in size, turn first yellow, and finally a more or less intense brownish red. Normal potassium chromate added to neutral solutions of silver causes the precipitation of normal silver chromate; but when the test drop is first acidified with nitric acid the crystals separating probably consist of both the chromate and bichro- mate. When recrystallized from hot nitric acid the precipitate will usually consist of the bichromate alone. When ammonium hydroxide is the solvent employed to recrystallize the silver chro- 336 ELEMENTARY CHEMICAL MICROSCOPY mate, the compound separating is thought to have the formula Ag2Cr0 4 - 4 NHg. 1 Normal potassium chromate produces in neutral or slightly acid solutions of manganous salts sheaves and bundles of a cinnamon brown manganous chromate soluble in excess of acid. Bichro- mates cause no precipitates in solutions of manganous salts. Precautions. The test drop must be moderately concentrated with respect to silver. When working with test drops acidified with nitric acid there is little danger of any interference by members of the calcium group. Large amounts of the salts of the alkalies seem to have an injurious effect when but little silver is present. In all analytical work it is safe to assume that the presence of any elements which are precipitated as chromate or bichromate in acid solution will interfere with the reaction for silver, particu- larly when such elements are in excess of the latter. White alloys believed to contain silver can be tested for this element by drawing across them a streak of a solution of ammo- nium bichromate in nitric acid. The color of the streak is gener- ally sufficient to indicate the presence or absence of silver, but if the streak of the reagent be examined under the microscope (best with an illuminating objective or some form of vertical illuminator) in the presence of silver the characteristic dark red crystals of silver bichromate will be easily distinguished. EXPERIMENTS. a. To a moderately concentrated neutral test drop add a fragment of (NH 4 )2Cr 2 O7. Then try K 2 CrO 4 . b. Acidify test drops with HNO 3 , then add the above reagents in turn. c. Decant the mother liquor from a precipitated test drop and recrystallize the Ag salt by heating with H^O. Try another preparation by heating with dilute HNO 3 . Recrystallize a third portion of the Ag compound, using NH 4 OH. d. Make a mixture of AgNO 3 and PbNO 3 , acidify with HNO 3 , then add a drop or two of dilute H 2 SO 4 and finally a fragment of (NH4) 2 Cr 2 O 7 . 1 Ladenburg, Handworterbuch, 10, 713. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 337 e. Repeat the last experiment, adding this time the (NH4) 2 Cr 2 O7 first, and then the H 2 S0 4 . /. Test several different preparations containing mixtures of the Ca group and Ag. g. Test a mixture of AgNO 3 and HgNO 3 . h. Make a rather concentrated neutral test drop of AgNO 3 , add a tiny crystal of Na 2 SO 4 . Study the Ag 2 SO 4 , which soon separates. Then add to the prepara- tion a fragment of (NH4) 2 Cr 2 O 7 . Note well all that takes place. If a selenate is at hand, substitute it in a new preparation for the Na 2 SO 4 . C. By Means of Arsenic Acid. The reagent is made by introducing into a drop of a dilute solution of arsenic acid a tiny drop of dilute ammonium hydroxide; stir. Apply the reagent by Method /, page 251. Silver arsenate AgsAsC^ (hexagonal) in the form of a fine granular precipitate is immediately produced; later, thin plates and plate-like prisms appear. The majority of the crystals which separate have the appearance of hexagonal plates. Their color by transmitted light varies from a reddish yellow in very thin plates to reddish brown with a tinge of dirty violet or even deep black as the thickness of the crystal increases. Crystallites bristling with long slender needles also abound. Silver arsenate is insoluble in acetic acid, soluble in hot nitric acid and easily soluble in ammonium hydroxide. Good prepa- rations can be obtained by recrystallizing from either of the latter solvents. In case ammonium hydroxide is employed, the colorless solu- tion resulting contains the compound Ag 3 As0 4 - 4 NH 3 , as has been shown by Widman. This tetra-ammonia salt can be made to crystallize in the absence of air in colorless needles, but on coming in contact with the oxygen of the air they turn red. It follows from this that the crystals obtained by recrystallizing silver arsenate from ammonium hydroxide are doubtless of vari- able composition. Although the crystals of silver arsenate are neat, well formed and characteristic, the reaction cannot be considered as a satis- factory one for silver because of the fact that most of the other metals usually associated with silver are also precipitated by 338 ELEMENTARY CHEMICAL MICROSCOPY arsenic acid, thus seriously interfering with the test. Solution of the precipitated arsenate in ammonium hydroxide and draw- ing off will usually effect a partial separation at least, and yield a more satisfactory test, but on the other hand the rendering of the drop alkaline may lead to the separation of arsenates which are soluble in acids but insoluble in alkaline solution. Arsenic acid applied as indicated may yield with calcium salts a separation of the compound NH 4 CaAsO4 6 H 2 0, ortho- rhombic, isomorphous with the corresponding phosphate; the crystals appear as large envelope-like crystallites with more or less ragged edges. If the solution be dilute hemimorphic forms identical with those of ammonium magnesium phosphate are seen, but generally of a larger size. Strontium yields minute stars and crystalline grains; barium a dense amorphous pre- cipitate. Members of the magnesium group yield colorless crystalline double ammonium arsenates isomorphous with their double ammonium phosphates. Good crystalline compounds will be obtained with the alkaline earths and with the magnesium group only when considerable ammonium hydroxide has been added to the reagent or when the test drop is distinctly ammoniacal; under these circumstances the detection of silver as arsenate may be masked. Although silver arsenate is of little value as an identity test for silver it is of considerable use in detecting arsenates. Precautions. The arsenic acid may be added directly to the test drop to either neutral or to weak nitric acid solutions, but the best and most uniform results seem to follow the procedure suggested above. The amount of ammonium hydroxide added to the reagent drop must never be sufficient to neutralize all the arsenic acid and give rise to an alkaline solution. Note. It is of theoretical interest to consider in connection with the arsenic acid test for silver, the behavior of compounds of the MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 339 elements analogous to arsenic as shown by their position in the Periodic System. We find, for example, crystalline salts of silver with phosphorus, as silver phosphate; with antimony, silver antimonate; with vanadium, silver vanadates; with chro- mium, silver chromates; with molybdenum, silver molybdates. Of these salts the chromates and vanadates can be employed for the detection of silver, but the phosphates, antimonates and molybdates cannot be made to yield sufficiently characteristic results. EXPERIMENTS. a. Test a neutral solution of AgNOs in the manner suggested above. b. Recrystallize a preparation of AgsAsO4 from HNOs. c. Try another preparation with NH 4 OH. d. Test a mixture of Ag and Pb. , Then one of Ag and Hg. e. Try the above reaction on salts of Ca, Sr and Ba, first alone, then in mix- tures but with no Ag present. /. Try salts of Mg, Zn and Cd. g. Try a salt of Ca in the presence of much NH 4 C1. COPPER. Crystal Forms and Optical Properties of Common Salts of Copper. A. ISOTROPIC. Cuprous chloride, bromide and iodide. B. ANISOTROPIC. Hexagonal. Tetragonal. Ammonium-copper chloride; potas- sium-copper chloride. Orthorhombic. Chloride; sulphate plus 4 NH 3 . Monoclinic. Acetate; potassium-copper sul- phate. Triclinic. Sulphate. DETECTION. A. By Means of Ammonium Mercuric Sulphocyanate. The reagent is applied by Method /, page 251, to neutral or weakly acid solutions ; it must be neither alkaline nor ammoniacal. The appearance, properties and peculiarities of copper mercuric 340 ELEMENTARY CHEMICAL MICROSCOPY sulphocyanate have been discussed at length under Zinc on page 309, to which the student is referred. To obtain the truly characteristic moss-like and radiating crystallites the drop being tested must contain but little copper. The double sulphocyanate is sufficiently soluble to require several minutes for its appearance in very dilute solution. Since the zinc salt is much less soluble and possesses the prop- erty of adsorbing any copper present with a change of color from white through brown and black, a little zinc acetate or sulphate added to the drop to be tested before the reagent is applied will greatly increase the delicacy of the reaction. Infinitesimal per- centages of copper may be thus detected. The sulphocyanate test is the most satisfactory and generally useful identity test for copper we possess. EXPERIMENTS. These have already been performed under Zinc. B. By Means of the Triple Nitrite Reaction. When copper alone is to be tested for, proceed as follows: To the moderately concentrated drop add a fragment or two of sodium acetate if free mineral acid is present, if not add a tiny drop of dilute acetic acid, next add a fragment of lead acetate and stir until dissolved. Finally add a fragment of potassium nitrite. The black triple nitrite of potassium, copper and lead K 2 CuPb(NO 2 )6 which is formed has been described under Lead, page 327 (q.v.). By adding rubidium, cesium or thallous salts the delicacy of the reaction may be greatly increased. If nickel is present it will separate as a triple nitrite of similar composition K 2 NiPb(NO 2 ) 6 , light yellow or yellow-brown, in squares and cubes of larger size. They differ from the copper compound in never being black. Cobalt is immediately precipitated as insoluble potassium cobalt nitrite. In testing alloys or mixtures likely to contain lead, copper, nickel and cobalt, it is best to modify the above procedure. Sodium acetate is first added, then potassium nitrite followed by MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 341 acetic acid. Cobalt will immediately be precipitated. If lead and nickel or copper are present the yellow or black or both triple nitrites will eventually separate. If none appears, a little lead acetate is added; tiny black squares and cubes indicate copper. Powerful oxidizing agents must be absent. EXPERIMENTS. a. Test for Cu in CuS0 4 ; in Cu(NO 3 ) 2 . b. Try the reaction in acid solution; in ammoniacal solution. c. Try in like manner a mixture of Cu and Ni, Cu and Co. C. Other Useful Reactions for Copper, which may arise in Testing for Other Elements. Cesium chloride forms two very characteristic double chlorides with copper CsCl - CuCl 2 in golden yellow rectangular plates, squares and short stout prisms and a less frequently met with orange colored salt of unknown formula. These char- acteristic double salts frequently appear when testing for tin, antimony, or bismuth with cesium chloride or on rare occasions when testing for aluminum. Ferric chloride also forms a yellow double chloride with cesium chloride. The color and the appear- ance of the cesium iron chloride is quite different from the copper salt and the combination does not take place so readily. Potassium ferrocyanide in acetic acid solutions yields an amorphous red-brown precipitate. Added to ammoniacal solu- tions there appear after a time white dendrites of copper ferro- cyanide ammonia 2 (NH 3 ) Cu 2 Fe(CN) 6 . 1 The addition of acetic acid causes these dendrites to become red. ALUMINUM. Crystal Forms and Optical Properties of Common Salts of Aluminum. A. ISOTROPIC. The alums (I). B. ANISOTROPIC. Hexagonal. Sulphate; chloride (6 H^O). Tetragonal. 1 Behrens, Anleitung, p. 75. 342 ELEMENTARY CHEMICAL MICROSCOPY Orthorhombic. Nitrate (usually M). Monoclinic. Nitrate (or O). Triclinic. DETECTION. A . By Means of Cesium Sulphate. Apply the reagent by Method ///, page 252. Cesium alum CsAl(SO 4 )2 12 H 2 O separates in large, beauti- fully formed, brilliant, colorless octahedra, dodecahedra or in combinations of the cube and octahedron (isometric). Dendrites and many faced crystal aggregates are also frequent. Test drops containing cesium alum have a great tendency to remain in a state of super saturation. Often a single large crystal only will appear. In such an event, crushing the crystal and drawing its fragments through the drop will almost invariably yield a large crop of well-formed crystals. Schoorl suggests keeping as a reagent a sample of pure cesium alum. When testing for aluminum he adds cesium sulphate (or chloride) and after concentration to about the point of super- saturation, the tiniest possible fragment of cesium alum is intro- duced into the preparation and instantly pressed upon and crushed with a platinum wire, thus seeding the drop and causing the immediate appearance of the alum crystals, providing of course that aluminum is present. Testing for aluminum with cesium sulphate leaves little to be desired as to accuracy and elegance, but requires a little practice to learn just the proper concentration. Too dilute a test drop requires very long waiting. Spontaneous evaporation leads almost invariably to supersaturation. Evaporation over the micro-flame is very unsatisfactory. On the other hand, the addition of the reagent to too concentrated a test drop gives rise to the immediate formation of dendritic masses and skeleton crystals. It is true that the experienced worker will usually at once recognize these dendrites as due to the presence of aluminum, but in view of the fact that beautiful and far more characteristic crystals can be obtained, the worker should not be satisfied with malformed crystals. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 343 In the presence of magnesium sulphate there is formed a double sulphate of magnesium and cesium; hence in dealing with such cases it is necessary to add a sufficient amount of cesium sulphate to permit of the formation of both the cesium magnesium sul- phate and the cesium alum. It is very seldom that the cesium magnesium sulphate separates; when it does the crystals are to be referred to the monoclinic system. Manganous sulphate will likewise form a double sulphate with cesium sulphate separating in monoclinic crystals. Cesium alum is one of a group of double sulphates known as " alums," having the general formula M 2 (SO4) 3 N2SO4 24 H 2 O, where M can be Al, Cr, Mn, Fe, In, Ga, Tl; and N- Na, K, Rb, Cs, NH 4 , Ag, or Tl. All alums are isomorphous, and are to be referred to the isometric system. Theoretically, therefore, one would be led to expect that the presence of ele- ments capable of taking the place of aluminum in alums would be liable to interfere with the test for aluminum. But in addition to their property of being able to replace aluminum in these double sulphates, we must consider the crystallizing power of the com- pounds formed. It is herein that lies the explanation of the value of cesium sulphate over and above that of any other of the sulphates we might be inclined to select. Of the above listed alum-forming elements, aluminum is the only one which unites with cesium or rubidium sulphates to form easily crystallizable alums. The other elements unite with these two sulphates only with difficulty, and the alums formed can be regarded, from a microchemical standpoint, as difficultly crystallizable. Sodium, potassium and ammonium sulphates readily unite to form more or less crystallizable alums with the other alum-forming elements as well. as with aluminum. Not infrequently it will be found that cesium alum has a marked tendency to adsorb various substances which may be present, leading to a modification of the crystal form or to colored solid solutions. Precautions. Although it is obvious that in the case of simple compounds converted into sulphates it is merely necessary to add the reagent 344 ELEMENTARY CHEMICAL MICROSCOPY and allow the preparation to crystallize, it is essential that due regard be paid to (i) just the right concentration, (2) the absence of much free sulphuric acid, (3) the absence of other free mineral or organic acids, (4) the absence of colloidal substances. To avoid most of these difficulties it is always advisable to proceed as follows: To the drop to be tested add ammonium hydroxide in slight excess, decant the solution and wash the gelatinous precipitate with water. Then add a drop of water and follow it with a very little dilute sulphuric acid, only just enough to dissolve the aluminum hydroxide. Warm gently; cool, and to the drop add a fragment of the reagent. After a few seconds, beautiful large crystals of cesium alum separate. Cesium chloride can be employed as reagent, providing that the solution to be tested contains a little free sulphuric acid. The chloride is, however, not as satisfactory as the sulphate, particu- larly in the hands of beginners, for cesium chloride crystallizes in the isometric system, thus sometimes leading to confusion. Cesium sulphate, on the contrary, crystallizes in the ortho- rhombic system. An examination of a preparation containing the latter salt, between crossed nicols, will therefore permit of an easy differentiation, between crystals of cesium sulphate and those of cesium alum. If cesium sulphate is not at hand it may be prepared from the chloride in this manner: Place a drop of sulphuric acid at the corner of a slide or on platinum foil; add a small crystal of cesium chloride and evaporate to dry ness. If no fumes of sul- phur trioxide escape, add another drop of acid and heat again. It is evident, that by this method of treatment, in the majority of cases, it is in reality primary cesium sulphate that is formed, and not the normal sulphate as implied above. Care must there- fore be exercised in its use. The difficulties often experienced with this test by the beginner are generally due to too much sulphuric acid in dissolving the aluminum hydroxide and to too much acid in preparing the cesium sulphate. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 345 EXPERIMENTS. a. To a test drop consisting of a solution of AkCSO^s add a fragment of the reagent. b. Precipitate another drop with NH 4 OH, decant, wash the precipitate, dis- solve in the least possible amount of H2SO4 and test. c. Try Rb2SO 4 as reagent; then K 2 SO 4 ; Na 2 SO 4 , (NHOaSOi. Try CsCl. d. Test for Al in the presence of free HC1; free HNO 3 . e. Test preparations containing Al and Fe; Al and Cr; Al and Mn; Al, Fe and Cr; Al and Mg; Al in the presence of phosphates. /. Prepare slides of chrome alum, iron alum, etc., then mixtures of these various alums; note isomorphism. B. By Means of Ammonium Fluoride. See Method XV, page 268. Apply the fluoride in solid form (Method///). Use a celluloid object slide. From neutral solutions or those containing at the most only a trace of free mineral acid a double fluoride separates having the formula 3 NEUF - A1F 3 or considering this to be an alumino- fluoride its formula may be written (NH^sAlFe. It crystallizes in very tiny clear-cut colorless octahedra belonging to the iso- metric system. Alumino-fluorides of the same formula of potassium, rubidium, cesium and sodium are known; they are even less soluble than that of ammonium and therefore can be obtained only in such minute crystals as to be useless as a test. Lithium alumino- fluoride is also very insoluble. The ammonium, potassium, rubidium and cesium salts are isometric and form isomorphous mixtures; but the sodium salt is monoclinic. In these alkali fluorine compounds the aluminum can be re- placed by titanium, chromium, iron and vanadium. But in the case of zircono-fluorides, silico-fluorides (see page 279) and plumbo-fluorides the salts have the composition M 2 RFe, where M is an alkali metal and R may be Zr, Si or Pb. Crystalline double fluorides of aluminum with copper, nickel and zinc have been described, but these are too soluble to appear under the conditions which usually obtain in an analysis. 346 ELEMENTARY CHEMICAL MICROSCOPY Precautions. Employ only neutral solutions. Always have an excess of ammonium fluoride, for if not a compound of different formula results appearing as very tiny rods, worthless as an identity test for aluminum. Salts of lithium, sodium and iron must be absent. The presence of silicon and analogous elements will generally seriously complicate matters, and may ruin the test, owing to the formation of silico-fluorides, etc. (See ammonium silico- fluoride tests, under sodium and barium.) Aluminum sili co- fluoride is gelatinous, and does not crystallize. Testing for aluminum with ammonium fluoride generally yields results a trifle quicker than Method A, but the delicacy of the reaction is very little greater. Moreover, Method B is subject to many complications and interferences, and there is always danger, in spite of great care, of damaging objectives by the corrosive vapors arising from the test drop, since objectives of moderate power and therefore short working distance must be employed. For these reasons, testing with ammonium fluoride cannot be considered as being as satisfactory as the cesium sulphate method. One of the chief reasons for inserting the test in this series is the fact that crystals of ammonium alumino- fluoride may occasionally appear when ammonium fluoride is being employed for other purposes, and the presence of alu- minum is not suspected. The method of testing for aluminum by heating with ammo- nium fluoride in a platinum cup has been described under Method XV, page 270 (q.v.). The results thus obtained are in most cases somewhat more reliable than those given above but require more time, patience and care. TIN. Crystal Forms and Optical Properties of the Common Salts of Tin. A. ISOTROPIC. Tetraiodide (I); potassium chloro- stannate (I). MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 347 B. ANISOTROPIC. Hexagonal. Tetragonal. Orthorhombic. Tetrabromide. Monoclinic. Stannous chloride + 2 H 2 0; s tan- nous fluoride, stannic chlorides. Triclinic. DETECTION. A . By Means of Cesium Chloride. Apply reagent by Method 7, page 251. In testing for tin it is best to evaporate to dryness repeatedly with moderately concentrated nitric acid, thus converting the element into the insoluble dioxide. The dry residue is extracted repeatedly with dilute nitric acid to remove interfering elements and finally dissolved in aqua regia and the excess of acid removed by evaporation. Dissolve the moist residue in water. There is thus obtained a compound which we may term chlorostannic acid, 1 with which cesium salts yield an immediate precipitate of cesium chlbrostannate C^SnCle in the form of tiny colorless highly refractive regular octahedra and cubes. Rubidium gives a similar compound of greater solubility and therefore yielding larger crystals, but of sufficiently high solubility to render the separation of the crystalline phase too slow to be of practical use. These three chlorostannates are isomorphous. The am- monium salt is more soluble than the above and the presence of ammonium compounds is therefore objectionable; the same is true of sodium which yields Na^SnCle 5 H 2 0. Both iron and copper are apt to be adsorbed by the tin oxide, in such an event yellow or red double chlorides of copper or iron and cesium will eventually make their appearance. Occasionally if much iron is present the crystals of cesium chlorostannate are colored yellow. Lead if present may give octahedra of cesium chloroplumbate 1 This compound may also be regarded as a hydrated stannic chloride. If evaporated to dryness there will be obtained SnCl 4 #H 2 O, where x is 3, 5 or 8. All three salts are crystalline and all can be referred to the monoclinic system. 348 ELEMENTARY CHEMICAL MICROSCOPY As already noted antimony gives hexagons and bismuth rhombs of the corresponding chloroantimonate and chlorobismuthate. In the event of no precipitate appearing after some time, add a fragment of potassium iodide. This may lead to the formation of cesium iodostannate C^Snle of less solubility than the chloro- stannate. The iodo-compound separates in yellow cubes and octahedra. 1 In the case of simple salts or mixtures it is usually sufficient to convert into chlorides by evaporating with hydrochloric acid; then dissolve in water, acidulate with hydrochloric acid and add the drop of cesium chloride solution. But in such an event one must remember that double chlorides of Sb, Bi, Cu, Fe, Al, Zn, Cd, Pb, etc., will almost invariably separate if present. If much tin is thought to be present use rubidium chloride in preference to cesium chloride. Note. It is of considerable theoretical interest to note that in the compounds of the type just considered M 2 RCl6, M 2 RBr 6 and M 2 RI 6 , M may be K, Rb, Cs, (NH*) and R may be Se, Te, Sb, Pb, Sn, Pt, Ir, Os, Pd, Ru. All salts of this series are iso- morphous (Groth). EXPERIMENTS. Defer until Bi is being studied. ARSENIC. Crystal Forms and Optical Properties of Common Salts of Arsenic. A . ISOTROPIC. Trioxide (I, also, but rarely mono- clinic) . B. ANISOTROPIC. Hexagonal. Triiodide; silver arsenate (second- ary, normal is I ?). Tetragonal. Secondary potassium arsenate. Orthorhombic. Calcium -ammonium arsenate; magnesium-ammonium arsenate. 1 It is probable that the product actually obtained is a solid solution of Cs 2 SnI 6 in CsoSnCk MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 349 Monodinic. Primary ammonium arsenate; pri- mary sodium arsenate. Tridinic. DETECTION. A. Through the Formation of Arsine and its Reaction with a Crystal of Silver Nitrate. Use the distilling tube, Fig. 133, page 248, as a generator, as indicated in Figs. 138 and 139. FIG. 139. Apparatus for the Detection of Arsenic. Fit the side tube with a plug of soft wood P. Introduce two or three fragments of arsenic-free zinc Z, and through a pipette dilute hydrochloric acid A (the acid will not flow into the lower part of the tube until the plug P is loosened). Insert a loose plug of absorbent cotton C which has been soaked in lead acetate and dried. The plug P is next withdrawn. The acid is allowed to flow upon the pure zinc; a tiny drop of water 5 is introduced into the side tube and the plug reinserted. This drop makes a tight seal and prevents loss of gas. The tube is now tipped downward and a tube drawn down to a capillary and containing loosely a tiny crystal S of silver nitrate and one L of lead acetate is attached by means of a short piece of rubber tube R. From time to time the crystal S is examined to see if it changes color. If after some minutes S remains clear and colorless remove P, insert the material to be tested by means of a bit of drawn-out 350 ELEMENTARY CHEMICAL MICROSCOPY glass tubing or a fragment of solid may be pushed in by means of a platinum wire. Close the tube by means of a drop of water and the plug P. The reaction may be hastened by warming A over the micro-flame. If arsenic is present the crystal of silver nitrate turns yellow due to the formation of a compound believed to have the composition AsAg 3 AgNOs, and rapidly changes to black through the reduction to metallic silver. The lead acetate remains unchanged unless hydrogen sulphide is evolved. In acid solution antimony will yield stibine which reacts upon silver nitrate in a similar manner although the yellow compound is practically never seen. Phosphine or hydrogen sulphide turn the silver nitrate black at once, but the sulphur compounds should have been held back by the lead acetate cotton. The crystal L is introduced merely to make sure that any blackening of S can- not be due to volatile sulphur compounds. To differentiate between arsenic and antimony we may sub- stitute fragments of aluminum for the zinc and a solution of potassium hydroxide for the acid. Under these conditions, no stibine is evolved, only arsine passes off with the hydrogen. Metallic antimony is precipitated in part and deposited in part upon the aluminum. In place of a crystal fragment of silver nitrate we may employ a fragment of mercuric bromide or a textile fiber soaked in mer- curic bromide and dried; in the latter case a much finer capillary tube can be used and the delicacy of the reaction is somewhat increased. Arsine turns mercuric bromide red or brown. B. By Reduction to Metallic Arsenic and Subsequent Oxidation to Arsenic Trioxide. The powdered material is mixed with a small quantity of dry anhydrous potassium ferrocyanide and introduced into a thin walled tube of hard glass drawn down to a point and fused. The tube is tapped gently to cause all the material to collect in the tip of the tube. Heat the material gently at first and finally raise the temperature to a red heat. The arsenical compound is reduced; arsenic is set free and condenses upon the walls of the tube as a brownish mirror. Antimony will yield a black or metal- lic mirror; mercury a sublimate of tiny silvery spheres. Certain compounds of carbon or sulphur may yield deposits upon the MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 351 glass closely resembling the arsenic mirror. It is therefore essential to carry the test a step farther; to this end, cut off the closed tip of the tube and heat the mirror over the micro-flame. The arsenic will be vaporized and oxidized, collecting upon the cool walls as A^Oa in the form of glistening colorless highly re- fractive (n = 1.755) isometric crystals in the form of octahedra or as derivatives of the octahedron. These crystals are soluble in potassium hydroxide solutions and are precipitated therefrom in the form of octahedra by strong nitric acid. ARSENATES. By Means of Silver Nitrate. Apply reagent by Method /, page 251, to the ammoniacal drop. This reaction has already been discussed at length under Silver, Method C, page 337. By Means of Zinc Chloride in Ammoniacal Solution. To the test drop add ammonium hydroxide, then apply the zinc chloride by Method 7, page 251. Ammonium zinc arsenate NKtZnAsC^ 6 H 2 O separates in the same forms as those described for ammonium magnesium phosphate (q. v.) with which it is isomorphous, as also with the compound NHiZnPC^ 6 H 2 O ; but the crystallites of the am- monium zinc arsenate are more feathery than those of the ammonium magnesium phosphate. Phosphates must be absent. ARSENITES. By Means of Silver Nitrate. Apply the reagent by Method /, page 251, to the ammoniacal drop. Lemon yellow silver arsenite is immediately precipitated first as an amorphous mass, later crystallizing in a variety of forms. The first crystals appear as exceedingly tiny acicular crystals in masses, stars and crosses, later as fusiform grains, and still later as thin rods with notched ends, or long irregular acicular prisms. Eventually some oxidation takes place and there will appear crystals of silver arsenate. Silver arsenite is soluble in acids and in ammonium hydroxide, hence the amorphous precipitate partially redissolves. 352 ELEMENTARY CHEMICAL MICROSCOPY EXPERIMENTS. a. Test by Method A the following: solutions of As 2 O 3 ; of NaAsO 2 ; of H^AsOaj one drop of commercial E^SO^ one drop of commercial HC1; trying first the AgNO 3 crystal and then the HgBr 2 fiber. b. Test the above compounds by Method B. c. Test the same compounds with AgNOa; and finally with ZnCl 2 . ANTIMONY. Crystal Forms and Optical Properties of Common Salts of Antimony. A. ISOTROPIC. B. ANISOTROPIC. Hexagonal. Red tri-iodide; strontium-antimonyl tartrate; lead-antimonyl tartrate. Tetragonal. Barium-antimonyl tartrate (T or 0). Orthorhombic. Yellow tri-iodide (O or M) ; ba- rium-antimonyl tartrate; potassium-anti- monyl tartrate; sodium-antimony 1 tartrate. Monoclinic. Antimonyl chloride. Tridinic. DETECTION. A . By Means of Cesium Chloride. Apply reagent by Method /, page 251, to the drop acidi- fied with hydrochloric acid. A double chloride of cesium and antimony of the formula 2 CsCl SbCls 2^ H 2 separates in hexagons and elongated six- sided plates. Many of the hexagons show a system of straight or curving ribs extending from the center to the angles of the hexagons. Bismuth yields rhombs or long plates showing an hexagonal outline, and having a lower solubility than the antimony salt. Copper yields a series of double chlorides varying in color from bright yellow to deep red. These salts usually separate in rec- tangular prisms, but the red compound sometimes assumes forms closely resembling the iodo-compounds referred to below. Tin causes the immediate precipitation of tiny regular octa- hedra of the formula C^SnCle, a salt of chlorostannic acid. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 353 Cesium chloride has remarkable powers of forming more or less difficultly soluble double chlorides with a large number of elements and we may thus expect to often find in preparations to which cesium chloride has been added an abundant crop of well-formed crystals, whose origin is puzzling unless we know what elements are present. The cesium chloride test is made more satisfactory and much more sensitive by obtaining an iodo-salt instead of that described above. This is accomplished by adding a fragment of potassium iodide to the test drop after applying the cesium chloride. Crys- tals of a double iodide of cesium and antimony having the same form as the double chloride are obtained but they are deep orange yellow or orange red instead of colorless. The composition of these crystals is not well established, but the weight of evidence seems to be that three molecules of Csl unite with two or three molecules of SbI 3 , rather than with SbLi. The test thus performed is an excellent one, but requires con- siderable experience in order to properly control the conditions. The test drop must be neither dilute nor concentrated and only just sufficient hydrochloric acid should be present to prevent an antimonyl compound from forming. It is also better to adopt for this iodide modification the method of applying the reagents suggested by Schoorl, 1 namely adding a fragment of cesium chloride to one side of the drop and a fragment of potassium iodide to the opposite side. The double iodide of bismuth separates in rhombs and elon- gated hexagons, rarely in the regularly formed hexagons of the antimony salt. Their color is a deeper orange (or even a red) than that of the antimony double iodide. Tin forms yellow cesium iodostannate in regular octahedra. Precautions. When iodine separates it is an indication that too small an amount of potassium iodide is present. In the event of a precipitate resulting upon the addition of hydrochloric acid at the beginning (Ag, Pb, Hg, Cu) sufficient acid should be added to complete the reaction. Decantation or 1 Beitrage z. mikrochem. Anal. Wiesbaden 1909, p. 49. 354 ELEMENTARY CHEMICAL MICROSCOPY filtration should then be resorted to and the clear solution care- fully concentrated to remove the excess of acid until a drop of water causes a precipitate of antimonyl (or bismuthyl) chloride. Then very carefully add hydrochloric acid with thorough stirring, until the precipitate just dissolves. EXPERIMENTS. Defer until Bi is being studied. ANTIMONATES. The composition of the various antimonates commercially available appears to be quite uncertain. The only one of im- portance is the potassium salt sold variously as potassium anti- monate, metantimonate or pyroantimonate ; it usually conforms fairly closely to the formula H 2 K 2 Sb 2 O7 6 H 2 0. It is difficultly soluble even in boiling water. Sodium salts in neutral solution yield, with antimonates of this type, very insoluble sodium pyroantimonate, separating as tiny lenticular grains or larger fusiform crystals singly or uniting in more or less globular masses. From dilute solutions what appear to be tetrahedra, octahedra or rectangular prisms are formed. Although appearing to be isometric the crystals are to be referred to the tetragonal system. Magnesium salts in neutral solution yield H 2 MgSb 2 07 9 H 2 O first as an amorphous precipitate, later crystallizing in thin transparent colorless hexagonal plates, and as small, irregular spherulites. Occasionally stars or rosettes or short hexagonal prisms are obtained. The magnesium salt is dimorphic, being either hexagonal or monoclinic according to conditions. Of the two tests that with sodium is the more satisfactory. If it is necessary to neutralize a test drop in testing for anti- monates use potassium carbonate. Ammonium salts interfere with the sodium and magnesium tests. BISMUTH. Crystal Forms and Optical Properties of Common Salts of Bismuth. A. ISOTROPIC. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 355 B. ANISOTROPIC. Hexagonal. Sulphate (9 H 2 O). Tetragonal. Bismuthyl chloride. Orthorhombic. Monodinic. Triclinic. Nitrate (?) ; bismuthyl nitrate. DETECTION. A . The Addition of Water to neutral or very faintly acid so- lutions followed by the formation of a heavy white amorphous or granular precipitate should lead to the suspicion of the presence of bismuth. From the chloride, the compound BiOCl is obtained and from the nitrate BiONOs H 2 0. B. By Means of Potassium Sulphate. This test has been discussed at length under Sodium, Method B, page 276, and also under Potassium, Method B, page 284, to which the student is referred for details. Neither arsenic, antimony, nor tin yield a crystalline deposit. The test is therefore one of the most satisfactory for the recog- nition of bismuth, providing lead is absent. Lead yields a gran- ular or amorphous (or rarely crystalline) precipitate with potas- sium sulphate. It is therefore necessary to first remove the lead by precipitating with sulphuric acid in the presence of nitric acid before proceeding to test for bismuth. With this end in view add to the solution to be tested nitric acid, then a drop of very dilute sulphuric acid if no precipitate results, evaporate until fumes of sulphur trioxide are formed. Then proceed as described under Method //, page 252, Experiment a. If a precipitate forms with the sulphuric acid decant, centrifuge or filter the solution to remove the lead, after which evaporate with sulphuric acid and proceed as above. C. By Means of Cesium Chloride. This test has already been discussed under Antimony, Method A, page 352. The only specific difference between the double chlorides of these two elements is that with bismuth there is a greater ten- dency toward rhombic plates. Conversion into double iodides, gives a salt darker colored than that with antimony. ELEMENTARY CHEMICAL MICROSCOPY A great excess of hydrochloric acid seriously reduces the delicacy of the reaction, while nitric and sulphuric usually pre- vent the separation of typical crystals. The student must bear in mind the caution given under anti- mony that cesium chloride has a strong tendency to form double salts, especially with lead, copper, cadmium, zinc, aluminum, etc. EXPERIMENTS. a. Try CsCl upon a solution of Sn in HC1. b. Try the test upon Sb in HC1 solution; upon Bi in HC1 solution. c. Try converting the chloro-salts of these three elements into the iodo com- pounds. d. Try testing for Sb and Bi in turn in the presence of a little Cu. e. Try mixtures in which some of the other metals are present which form crystallizable double chlorides with CsCl. D. Other Important Tests. With Primary Potassium Oxalate. (See Manganese, IMethod A, page 359.) With Ammonium Bichromate. (See Silver, Method J5, page 334.) CHROMIUM. Crystal Forms and Optical Properties of Common Salts of Chromium. A. ISOTROPIC. Chrome alums (I). B. ANISOTROPIC. Hexagonal. Tetragonal. Orthorhombic. Barium chromate (or M) ; calcium chromate (or M); potassium chromate; silver chromate; sodium chromate; stron- tium chromate (or M) ; zinc chromate. 1 Monoclinic. Ammonium bichromate; ammonium chromate; barium chromate (or O) ; calcium chromate (or O) ; lead chromate ; strontium chromate (or O). Triclinic. Potassium bichromate; silver bichro- mate; 2 sodium bichromate. 1 In the presence of FeSO4 7 H^O the salt separates monoclinic. 2 Ag 2 Cr 2 O 7 dissolved in water decomposes into Ag 2 CrO 4 and CrO 3 . MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 357 DETECTION. A. In simple salts we may obtain the following colors and reactions: a. Soluble chromates are yellow, bichromates red, their solutions yellow. Solutions of chromium salts where chromium acts as a base, when heated in acid solution, are green. b. Chromium yields with ammonium hydroxide a bluish or greyish green or greyish lavender hydroxide. In the presence of ammonium salts, especially ammonium chloride, this hydrox- ide is partially soluble with the formation of the compound CrCla 4 NH 3 . Boiling drives off the ammonia and chromium is completely precipitated as Cr(OH) 3 . c. Silver nitrate gives in solutions weakly acid with nitric acid a characteristic deep red chromate with both chromates and bichromates (see Silver, Method B, page' 334). In neutral solutions the bichromate yields a crystalline silver chromate somewhat more readily than the bichromate but the difference is too slight to be of any practical use in differentiating between the salts. d. Alkali chromates added to neutral solutions of manganous salts give a characteristic manganous chromate, but alkali bi- chromates give no such reaction (see Manganese, Method J5, page 360). B. By Conversion into Cesium Chrome Alum. To a drop of the solution to be tested add ammonium hydroxide. Should a reddish liquid result, boil. Decant the solution from the bluish or greenish precipitate. Wash the pre- cipitate once or twice. -Add a tiny drop of water and then very carefully the least possible amount of dilute sulphuric acid which will just dissolve the precipitate. Evaporate carefully nearly to dryness and add a tiny drop of water. Finally introduce near the center of the drop a fragment of cesium sulphate. Cesium chrome alum will almost immediately separate in characteristic alum crystals, the octahedron and dodecahedron predominating (isometric). These crystals have a faint bluish tint by trans- mitted light. The peculiar purple color of chrome alum will not be seen unless they attain a relatively large size and reflec- 358 ELEMENTARY CHEMICAL MICROSCOPY tions from their faces become noticeable. To be of value as a test for chromium both the crystal form and color must be taken into account. Free mineral acids should be absent, as also the salts of organic acids. In general we must observe the same precautions as in test- ing for aluminum with cesium sulphate. (See Aluminum, Method A, page 342.) It is obvious that other "alum" forming elements, such as aluminum, iron and manganese, must be absent or present only in traces. Since all the alums are isomorphous it is often possible to start crystallization by introducing into the test drop an infinitesimal trace of potash alum when by chance the preparation shows a tendency to supersaturate and no crystals form, or even better, add a similar tiny fragment of cesium alum. In such an event we must place our chief dependence upon the color of the crystals separating. EXPERIMENTS. Test simple salts of Cr as described above, then employ more or less complex mixtures. C. Detection of Chromium in Complex Mixtures such as Alloys, etc. Method of Behrens. 1 Place the finely-divided material on an object slide. Add a fair-sized drop of concentrated nitric acid, heat to boiling, decant the acid to another slide and treat the residue again in the same manner. Repeat until all is dis- solved or until sufficient material has passed into solution. Unite all the drops and evaporate to dryness. By means of a tiny spatula carefully scrape off the dry mass into a platinum cup or upon a piece of platinum foil. Add a very small quantity of sodium carbonate-potassium nitrate fusing mixture (3:1) and heat until a clear fusion results, adding more fusing mixture if necessary, but being careful to use no more than absolutely neces- sary. The yellow fused mass is dissolved in water, concentrated 1 Behrens, Anleitung, i Auf. 186. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 359 to small bulk, acidified with acetic acid, a trace of sulphuric acid added and into the drop, a drop of silver nitrate is caused to flow. Silver sulphate will first separate in its characteristic form but will be colored yellow or red through the solid solution of silver chromate in it. Later the red-brown or blackish crys- tals of silver chromate appear. EXPERIMENTS. a. Look over notebook records of experiments made under Silver Exps., Method B, page 336. Similar crystals will be obtained upon testing for Cr with AgN0 3 . b. Test for Cr in several different Cr compounds by Method B. c. Test by Method B in Cr salts, mixed with Al, Fe, Cu, Ni. d. Test for Cr in chrome iron. MANGANESE. Crystal Forms and Optical Properties of Common Salts of Manganese. A. ISOTROPIC. B. ANISOTROPIC. Hexagonal. Tetragonal. Orthorhombic. Potassium permanganate. Monoclinic. Acetate (ous) ; chloride (ous) ; am- monium-manganous sulphate; potassium- manganous sulphate ; sodium-manganous sulphate. Triclinic. Sulphate (ous). DETECTION. A. With Manganous Salts Oxalic Acid or Primary Potassium Oxalate forms Characteristic Crystals of Manganous Oxalate. Obtain a thin uniform film of dry potassium oxalate upon the slide; Method IV, page 255. Draw across this film the neutral solution of the material to be tested or a solution slightly acidified with acetic acid. Six-armed stars of MnC 2 O 4 3 H 2 O separate. These stars result from the intersection of thin twinned prisms. They polarize strongly, extinguish parallel to their length and exhibit brilliant polarization colors. 360 ELEMENTARY CHEMICAL MICROSCOPY This test is excellent when pure manganous salts are being dealt with, but is seriously affected by much alkali and ammonium salts or by the presence of those elements readily precipitated as oxalate, for example, the elements of Group VIII of the Periodic System, or those of Group II. Free mineral acids seriously interfere. With solutions highly concentrated with respect to manganese no reaction will be obtained nor will satisfactory results follow the use of too dilute test drops. Silver, lead, mercurous and stannous salts should be absent. EXPERIMENTS. a. Test as above MnSO 4 . Then try the addition of a drop of H 2 C 2 O 4 to a test drop by Method /, page 251. b. Try effects of free acids upon the test. c. Test mixtures of MnSO 4 with members of Group VIII. B. By Means of Potassium Chr ornate. Apply reagent to test drop by Method ///, page 252. Sheaves of yellowish brown, acicular, strongly pleochroic crystals separate from neutral or feebly acid solutions; but from drops containing a trace of free nitric acid stout dendritic masses and clusters of yellowish brown prisms are obtained. The test drop should be moderately concentrated. Nitric acid greatly slows down the reaction and if present in more than traces prevents the formation of crystals. The other mineral acids behave in a similar fashion. With pure manganous salts this test is excellent, but is of little value in the presence of silver, lead, mercury or in fact any element forming a difficultly soluble chroma te. See Silver, Method JB, page 334; Mercury, Method B, page 321. Potassium bichromate applied as above gives no crystalline precipitate. EXPERIMENTS. a. Test a drop of MnSO 4 with K 2 CrO 4 ; with K 2 Cr 2 O 7 . b. Repeat the test, previously acidifying with HNO 3 ; with HC1; with HC 2 H 3 O 2 . c. Repeat in the presence of Ag, of Pb. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 361 C. Through Fusion with a Mixture of Sodium Carbonate and Potassium Nitrate. The fusion should be made in a small platinum cup or upon platinum foil, using the smallest possible amount of the fusing mixture which will react with the unknown. It is always wise to first obtain the hydroxide or oxide and employ this material for the fusion. If manganese is present a green color is obtained, due to the formation of manganates of sodium and potassium Na2MnC>4, K 2 MnO 4 . Iron and chromium mask the reaction. EXPERIMENTS. a. Test several different Mn compounds by fusing on platinum foil or in a bead on Pt wire. D. By Means of Phosphates in Ammoniacal Solution. Manganous salts are precipitated as NH 4 MnPO4 6 H 2 O. See Magnesium, Method B, page 304; Nickel, Method B, page 365; Cobalt, Method C, page 367. The hemimorphic crystals obtained usually grow somewhat longer than those of magnesium but are otherwise identical. They are proved to be due to manganese by adding hydrogen peroxide which causes them to turn brown. E. By Means of Sodium Bismuthate. Dissolve the material in concentrated nitric acid and evaporate the solution to dryness. Dissolve in dilute nitric acid, add several small portions of sodium bismuthate, stirring after each addition, allow to stand a short time; a pink or purple color results with a precipitation of brown oxide of manganese. Next add very carefully in tiny fragments just sufficient, no more, sodium thiosulphate to dissolve the precipitated oxide. A colorless milky drop results; add a drop of nitric acid (1:4) and stir thoroughly. Now again add carefully and slowly a very little at a time sodium bismuthate. A beautiful pink or purple color is developed due to the permanganate formed. 362 ELEMENTARY CHEMICAL MICROSCOPY To complete the test add a fragment of rubidium chloride, stir, add a drop of water and allow a drop of perchloric acid to flow into the drop. Crystals of rubidium perchlorate are imme- diately formed, taking up the permanganate in solid solution and yielding pink or purple crystals. EXPERIMENTS. Test this method first upon pure Mn salts, then upon mixtures of other elements with Mn. IRON. Crystal Forms and Optical Properties of Common Salts of Iron. A, ISOTROPIC. Iron alums (I). B. AN ISOTROPIC. Hexagonal. Chloride (when sublimed). Tetragonal. Orthorhombic. Ammonium-ferric chloride; oxa- late (ous). Monoclinic. Sulphate (ous); 1 ammonium-ferrous sulphate; sodium-ferric oxalate ; potassium- ferric oxalate. Triclinic. DETECTION. A . By Means of Potassium Ferrocyanide. To the test drop, apply a fragment of the reagent by Method ///, page 252. A dark blue precipitate or color indicates iron. The precip- itate is soluble in alkalies, insoluble in acids. It is therefore always best to acidify with hydrochloric acid before adding the ferrocyanide. The presence of much copper may seriously interfere with the test because of the formation of brown copper ferrocyanide. 1 But if magnesium sulphate is present, orthorhombic. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 363 EXPERIMENTS. a. Test for Fe in simple salts. b. Test in complex mixtures with other elements which will be precipitated by K 4 Fe(CN) 6 . NICKEL. Crystal Forms and Optical Properties of Common Salts of Nickel. A. ISOTROPIC. Ammonia nickel nitrate (I). B. ANISOTROPIC. Hexagonal. Tetragonal. Orthorhombic. Sulphate. Monoclinic. Acetate; chloride; nitrate; sulphate; ammonium -nickel sulphate; potassium- nickel sulphate. Triclinic. DETECTION. CH 3 - C = NOH A. By Means of Dimethyl Glyoxime, I CH 3 - C = NOH To a drop of the solution to be tested add ammonium hydroxide until in slight excess. Decant the solution of the hydroxides which have been dissolved by the ammonium hydroxide, from those which are insoluble. Close to the clear ammoniacal drop place a large drop of a freshly prepared saturated solution of dimethyl glyoxime. Cause the ammoniacal drop to flow into the reagent. Nickel yields an immediate rose-pink or magenta-colored precipitate at first amorphous in character, later changing into a felt of exceedingly fine acicular crystals. Near the edges of the crystalline mass tiny needles form in star-like and irregular bristling clusters. Often a yellow precipitate is first formed, changing only slowly into pink. The nickel salt of dimethyl glyoxime has the formula 364 ELEMENTARY CHEMICAL MICROSCOPY No other element yields a similar appearing compound. The reaction is an exceptionally sensitive one; exceedingly small amounts of nickel may be thus detected save in the pres- ence of large amounts of cobalt or copper. Neither cobalt 1 nor copper alone yield a precipitate, but both these metals mask or prevent the formation of the typical nickel compound; a yellow amorphous precipitate results in which can be found only a few masses of the pink needles. Copper can be easily removed by deposition upon a piece of zinc foil prior to the addition of the ammonium hydroxide. This is accomplished by placing the weakly acid drop upon a clean bright piece of zinc. As soon as a black spot is formed the drop is decanted to a new position, and as soon as it is ob- served that the zinc is not at once stained the drop is decanted upon an object slide, ammonium hydroxide added and the test for nickel applied. Cobalt may be removed by adding to the almost neutral drop, a fragment or two of potassium nitrite, warming to hasten solu- tion, and then adding a drop of acetic acid. Potassium cobalt nitrite is precipitated. After a few seconds the liquid is de- canted from the precipitate which clings tenaciously to the glass and ammonium hydroxide is added, ignoring any few tiny par- ticles of the nitrite which may have been carried over. The glyoxime test can now be applied with assurance of detecting nickel if present. An excess of neither silver nor zinc appears to influence the reaction for nickel. Dimethyl glyoxime gives with iron salts a red color. In testing for nickel, therefore, we often obtain an indication of the presence of iron in spite of the fact that ammonium hydroxide has been added; for in the presence of ammonium salts the addition of ammonium hydroxide to ferrous solutions will not precipitate all the iron, owing to the formation of soluble double 1 All the samples of cobalt salts sold as C.P. tested by the author have given a slight precipitate with the reagent, probably due to traces of nickel present in the material. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 365 salts, such as (NH 4 ) 2 SO 4 - FeS0 4 or 2 NH 4 C1 FeCl 2 . Non- volatile organic acids prevent the precipitation of ferric hydroxide and the ferric salts thus remaining in solution will react with the glyoxime. B. Other Tests for Nickel 1 . Triple nitrite of lead nickel and potassium K 2 PbNi(N0 2 )e. See Lead, Method C, page 327; Copper, Method B, page 340. 2. Ammonium nickelous phosphate NH 4 NiPO 4 6 H 2 O. See Magnesium, Method B, page 304. This salt is isomorphous with the magnesium salt. Note. The addition of hydrogen peroxide causes no change in the color of the crystals of ammonium nickel phosphate, but will turn those of cobalt brown. EXPERIMENTS. a. Try the glyoxime reaction on salts of Ni in NH 4 OH and in acid solution; and in different concentrations. 6. Try test upon Co compounds. c. Make a mixture of Ni and Co and test. d. Test for Ni in the presence of much Cu. Remove the Cu from a drop by means of metallic Zn and test again. Then try the detection of Ni in the presence of much Fe. e. Apply the phosphate test to a Ni salt and as soon as the crystals are well formed, allow a drop of H 2 O 2 to flow into the drop. Repeat the process with a Co salt. COBALT. Crystal Forms and Optical Properties of Common Salts of Cobalt. A. ISOTROPIC. B. ANISOTROPIC. Hexagonal. Tetragonal. Orthorhombic. - - Ammonium-cobalt phosphate; purpureo-chloride (pseudo tetragonal) . Monoclinic. Acetate; chloride; luteo-chloride; nitrate; potassium-cobalt sulphate; roseo- chloride; sulphate. triclinic. 366 ELEMENTARY CHEMICAL MICROSCOPY DETECTION. A. By Means of Ammonium Mercuric Sulphocyanate. See Zinc, Method A, page 307; Copper, Method A, page 339- Mercury cobalt sulphocyanate Hg(CNS) 2 Co(CNS) 2 sepa- rates as dark blue prisms, usually in irregular clusters. Its solutions have the tendency to supersaturate and it is therefore necessary to give the reaction considerable time, or even evapo- ration over the micro-flame may be advisable. Crushing the first crystals appearing near the circumference of the drop and drawing the fragments across often expedites the reaction. Nickel yields no crystals and does not interfere unless in ex- cessively great amount. Precautions. The test drop should be neutral or only slightly acid with acetic acid, but must not be alkaline. Better results are to be obtained with mineral acid salts than with those of organic acids. EXPERIMENTS. The student should refer to his notes under Zn, where the results of his ex- perience with the reagent upon Co should be found. B. By Means of Potassium Nitrite. To the neutral or slightly acid drop add a fragment of potassium nitrite. Stir. Then warm and add a drop of acetic acid. Potassium cobalt nitrite 3 KN0 2 Co(N0 2 ) 3 i| H 2 is im- mediately precipitated in the form of tiny cubes, so minute as to simulate an amorphous or finely-granular deposit. These crystals appear black by transmitted light, yellow by reflected light. From hot solutions there may sometimes be obtained crystals recognizable as cubes and octahedra. This test has its greatest value in a negative way since failure to obtain the very insoluble double nitrite may be considered as indicative of the absence of cobalt. MICROCHEMICAL REACTIONS OF THE COMMON ELEMENTS 367 Upon obtaining a yellow precipitate, decant the supernatant liquid, convert the double nitrite into the chloride, nitrate or sulphate and test for cobalt by Method A . EXPERIMENTS. These have already been tried under Lead, Method C, page 329 (q.v.). C. Other Tests for Cobalt. As ammonium cobaltous phosphate NELiCoPC^ 6 H 2 ; isomorphous with the magnesium, nickel x and manganese am- monium phosphates. See Magnesium, Method J5, page 304. Add hydrogen peroxide and warm. The cobalt compound turns brown. THE QUALITATIVE ANALYSIS OF MATERIAL OF UNKNOWN BUT OF SIMPLE COMPOSITION. The following brief outline may serve as a guide to the steps to be taken in the microchemical analysis of the simple " un- knowns" which will be given to the student for practice. 1. Examine the material with a low power. 2. If found to consist of several components, try to isolate them, using forceps, or scraping off particles with a knife point, a file or a tiny drill. 3. Test solubilities in water, HNOs, and NELiOH and note whether the solutions obtained yield crystals on evaporation. 4. Subject the crystalline material to polarized light and deter- mine whether it is isotropic or anisotropic. If the latter, whether it exhibits parallel or oblique extinction. This should afford a clue as to the probable nature of the salts obtained. 5. Add a drop of dilute HC1 presence or absence of Pb, Ag, Hg, (Cu). Decant, and test solution as in 6. To residue add NH 4 OH. 6. Add a drop of dilute H 2 SO 4 presence or absence of Ca, Sr, Ba, Pb, Ag, Hg, (Sb, Bi), and salts of low solubility. 7. Add to a drop, a drop of ammonium mercuric sulpho- cyanate presence or absence of Zn, Cu, Cd, Co, Fe, (Ag), (Pb), (Au), (Mn). 368 ELEMENTARY CHEMICAL MICROSCOPY 8. Add to a drop a little CsCl and KI presence or absence ofPb,Sn,Hg,Ag,Sb,Bi,(Cu). 9. Add to HC1 solution H 2 PtCl6 presence or absence of K, NH4, Na. 10. Add zinc-sulphide-fiber, then at once a drop of NELtOH - presence or absence of Pb, Ag, Hg, Cu, Bi, Cd, As, Sb, Sn, Ni, Co, Fe, Al, Cr, Mn, Au, and salts insoluble in NELtOH. 11. Or instead of proceeding as in 10, add HKC 2 O 4 pres- ence or absence of Ca, Sr, Ba, (Mg), Zn, Cd, Sb, Sn, Pb, U, Mn, Fe, Ni, Co, Cu, Ag, (Hg), (Cr). 12. Test for the acids. 13. From the solubility tests, optical behavior and above reactions make a list of the elements which are probably absent and proceed to make identity tests for those which can be present. In all tests applied observe carefully the precautions, notes and interferences given in the discussion of the test. 14. Make special tests for silicates, titanates, etc. THE COMMON ACIDS. The detection of the acid radicals (anions) by microchemical reactions is much more difficult than the identification of the bases (cations). This is largely due to the relatively high concentrations employed, to rapid evaporation taking place during manipulations and to the fact that through the addition of reagents many complex salts are formed of lower solubilities than those originally existing in either unknown or reagent. In the elementary course whose outline is covered by this textbook the identification of the acid radicals in simple salts or simple mixtures alone is undertaken. With materials of this nature the qualitative analysis is comparatively easy and no elaborate directions or schemes of procedure are necessary. Most of the tests for the acids have already been studied and it is merely necessary in most cases to reverse the test for the bases to enable us to properly identify the acids. The behavior of the crystals, obtained in a test, toward polar- ized light will be found to be of great value in identifying the MICROCHEMICAL REACTIONS OF THE COMMON ACIDS 369 salts present in a mixture. The student should have acquired therefore, early in the course, the habit of examining his prep- arations between crossed nicols. Proceeding in this manner in connection with the qualitative tests we can usually determine the true nature of the salts present. In testing for the acids it is essential that the student shall always examine the preparations before they evaporate to dry- ness and that he shall carefully observe the various precautions which have been given in the discussion of the various tests for the bases. When dealing with an unknown substance first spread out a little of the dry material upon a slide and examine it with a low power. If the material is not homogeneous, endeavor to pick out particles of its different components, using a plat- inum wire or glass rod. Then work upon each component separately. Try the solubility in water, acids, etc. Test the reaction toward litmus-silk (Method VIII, page 260) or other indicator. If the material is crystallizable, make observations as to its probable crystal system. Test the crystals between crossed nicols. Finally make rough estimations of the refractive indices by the immersion method or make melting-point determinations, or both, if possible. For convenience in microchemically testing for the acids we may make use of the following slight modification of the Bunsen- Treadwell classification of the acids, based upon the behavior of their salts toward silver nitrate, and toward barium chloride, in neutral and in nitric acid solutions. In case a free acid is to be dealt with it is best to add ammo- nium hydroxide in slight excess and drive off the excess, after neutralization, by evaporation to dryness. Then proceed as follows : I. To a drop of the moderately concentrated aqueous solution of the un- known apply a drop of concentrated solution of silver nitrate by Method /, page 251. 370 ELEMENTARY CHEMICAL MICROSCOPY A. No precipitate is produced and no crystalline deposit is ob- tained until the drop concentrates through spontaneous evapo- ration. See I. A, below. B. A colored precipitate is produced. See I. B, below. C. A white or wlorless precipitate is produced. See page 371. After a few seconds apply a small drop of nitric acid (i : 3) to the zone of precipitate. 1. The precipitate dissolves in whole or in part. If only in part, decant the solution and apply a fresh drop of nitric acid to the residue, to ascertain if the unknown consists of a mixture of both soluble and insoluble silver salts. 2. The precipitate is unaffected. n. To another drop of the dilute aqueous solution add a drop of barium chloride solution. See page 251. A. No precipitate results. See page 372. B. An amorphous, granular or crystalline precipitate is pro- duced. See page 372. 1 . The precipitate is soluble in whole or in part in nitric acid. 2. The precipitate is insoluble in nitric acid. III. To a drop of the dilute aqueous solution of the unknown material add a drop of nitric acid. A granular or amorphous precipitate results. See page 373- I. A. No Precipitate with Silver Nitrate. Chlorate. Fluoride; silicofluoride. 1 Nitrate. Perchlorate. 1 Sulphate. 1 I. B. The Precipitate is Colored (by Reflected Light). Arsenate. Red, brown or thick crystals black. Arsenite. Yellow. Chromate, bichromate. Red, brown or black. 1 Crystals separate slowly from moderately concentrated solutions or even from dilute solutions on long standing. MICROCHEMICAL REACTIONS OF THE COMMON ACIDS 371 Ferricyanide. Yellowish-red, or brownish-red. Iodide. So faintly yellow as to appear white, lodate. So faintly yellow as to appear white. Manganate, permanganate. Violet. Nitrite. Colorless unless in masses, then greenish. Phosphate. Yellow. Sulphide. Black or brown. I. C.i. The White or Colorless Precipitate Dissolves. ~ , Appearance of the precipitate before the nitric acid is applied. Acetates. Crystalline; prisms and plates. Borates. Granular. Carbonates. Amorphous or granular. Cyanates. Dense amorphous. lodates. Granular or crystalline in tiny stars or fine needles. Difficultly soluble in HNO 3 . Nitrites. Long slender needles. Oxalates. Granular or crystalline; short stout prisms, rhombs or hexagons. Sulphates. Prisms, rhombs and crystallites. Sulphites. Granular or crystalline; prisms. Tartrates. Amorphous becoming crystalline; crystallites and prisms. Thiosulphates. Dense amorphous, or granular, white changing to yellow, red-brown or dark brown due to formation of silver sulphide. When much sulphur separates the precipitate may ap- pear to be insoluble in HN0 3 . I. C. 2. The Colorless Silver Salt is Insoluble in Nitric Acid. Chloride. Bromide. Iodide. Hypochlorite. Ferrocyanide. 1 Sulphocyanate. 1 Turns yellowish red or brown when drop of nitric acid is applied. 372 ELEMENTARY CHEMICAL MICROSCOPY II. A. No Immediate Precipitate is Obtained with Barium Chloride. Acetate. Arsenate. 1 Ferrocyanide. 1 Borate. 1 Iodide. Bromide. Nitrate. Chlorate. Nitrite. Chloride. Oxalate. 1 Cyanide. Cyanate. Ferricyanide. II. B. i . Barium Chloride gives a Precipitate Soluble in Nitric Acid. 2 ggUg Appearance of the precipitate before the nitric acid is applied. Arsenites. Amorphous. Carbonates. Amorphous or granular; becoming crystalline. Chromates, bichromates. Yellow granular, or crystalline, only slowly soluble in nitric acid. Cyanates. From concentrated solutions, in prisms. Fluorides. Granular, lodates. Stars and dendrites. Only slowly soluble. Phosphates. Amorphous or granular. Sulphites. Granular or crystalline. Tartrates. Granular. II. B. 2. The Precipitate obtained with Barium Chloride is Insoluble in Nitric Acid. Silicon* uoride. Sulphate. Chromate, bichromate and iodate precipitates are only slowly soluble in nitric acid. 1 With concentrated solutions of these salts barium chloride will give a slowly formed crystal deposit. 2 Concentrated nitric acid precipitates barium nitrate in large colorless, iso- metric crystals. MICROCHEMICAL REACTIONS OF THE COMMON ACIDS 373 III. Nitric Acid produces an Amorphous or Granular Pre- cipitate. Molybdate. Silicate. Tungstate. Titanate. Zirconate. Note. It must be remembered that the addition of strong nitric acid will cause a crystalline precipitate in the case of many salts of low solubility. A somewhat better scheme of separation of the acids has been proposed by C. G. Hinrichs 1 based upon the behavior of their salts toward acetic and sulphuric acids when heated. Group I. Salts which when heated with strong acetic acid are decomposed and certain components are volatilized. Carbonate (CO 2 ). Cyanide (HCN). Hypochlorite (to Cl). Hyposulphite (S0 2 ). Nitrite (oxides of N). Sulphide (H 2 S). Sulphite (SO 2 ). Group II. Salts which when heated with strong sulphuric acid are decomposed and certain components are volatilized. Acetate (HC 2 H 3 O 2 ). Borate (B(OH)). Bromide (HBr). Chlorate (HC10 3 ). Chloride (HC1). Cyanate (CO 2 and NH 3 , latter forms (NH0 2 SO 4 ). Ferrocyanide (HCN). Ferricyanide (HCN). Iodide (HI). Nitrate (HN0 3 ). 1 Hinrichs, Microchemical Analysis, p. 116, St. Louis, 1904. 374 ELEMENTARY CHEMICAL MICROSCOPY Group HI. Non- volatile with sulphuric acid. Arsenate. Arsenite. Chromate, bichromate. Manganate. Permanganate. Phosphate. Sulphate. The separation by the above method may be carried out as described under Distillation, page 245. ACETATES. a. With Silver Nitrate in concentrated, approximately neutral solution, pearly scale-like crystals of silver acetate are obtained. Later these develop into long thin prisms with more or less Irregular sides and ends. Those in which six edges are de- veloped give terminal angles a trifle over 90 degrees, and extinc- tion almost parallel with their length (extinction angle 8 degrees). To confirm the test distil a portion acidified with phosphoric acid. In the absence of phosphoric acid, sulphuric acid may be employed. b. With Mercurous Nitrate added to concentrated solutions. Colorless plates and prisms; the thin six-sided prisms have their terminal angles equal to 100 degrees and exhibit parallel ex- tinction. c. With Sodium Chloride and Uranyl Nitrate in approximately neutral solutions. Sodium uranyl acetate is obtained. See Sodium, Method A, page 274. Add the uranyl nitrate to the drop of unknown, and draw this solution across the dry film of sodium chloride. ARSENATES. a. With Silver Nitrate. See Silver, page 337 ; Arsenic, page 351 . b. With Zinc Acetate and Ammonium Chloride in Ammoniacal Solution. See Magnesium, page 306. ARSENITES. a. With Silver Nitrate. See Arsenic, page 351. MICROCHEMICAL REACTIONS OF THE COMMON ACIDS 375 BORATES. a. With Ammonium Fluoride in Dilute Hydrochloric Acid Solu- tion. Add to the drop on a celluloid slip NaCl, or BaCl 2 , then the reagent, then a trace of HC1. See Sodium, page 279. Precautions. Silicon, titanium and zirconium must be ab- sent. The test drop must be moderately concentrated. b. Test with a Turmeric Linen Fiber. See page 261. BROMIDES. a. Staining Starch Yellow. To a drop of the solution to be tested add a trace of dilute sulphuric acid, warm gently. Cool. Add a very little potato starch, just enough to give a few granules in the center of the drop. Introduce at the center of the drop a small crystal of ammonium persulphate. Bromine is set free and colors the starch granules yellow. If iodides are present the starch will be colored blue or violet. Too long and too high heating will result in the loss of hydro- bromic acid. The preparation must be cool when the starch is added, otherwise the granules will be destroyed. The preparation must be examined at once, otherwise the yellow color will have disappeared. b. Silver bromide (and silver chloride) is soluble in ammonium hydroxide; silver iodide is not. CARBONATES. a. Characterized by Effervescence with hydrochloric or sulphuric acid. Gas bubbles visible in gelatin. See page 263. Cyanates give a similar reaction, carbon dioxide being formed by the reac- tion between cyanate and acid. b. In Solutions of Carbonates, Lead Acetate produces character- istic crystals of lead carbonate. c. To test the character of the gas given off, place in the distilling apparatus, Fig. 130, page 245, exposing a drop of lead acetate to the vapors. 376 ELEMENTARY CHEMICAL MICROSCOPY CHLORIDES. a. With Silver Nitrate. See Silver, page 331. b. With Lead Nitrate. See Lead, 'page 325, CHLORATES. a. Test the material with Rubidium Chloride and a little Potas- sium Permanganate to be sure perchlorates are absent (see Ex- periment a, Method IX, page 262). Then Convert into Perchlo- rates as follows: Dissolve a little of the material in a drop of water at the corner of an object slide, evaporate to dryness. Add a drop of sul- phuric acid, evaporate to dryness and heat until white fumes escape. Add a second drop of acid and heat until the excess of sulphuric acid has been driven off. Cool. Add a tiny drop of potassium permanganate (just sufficient to color the drop) and a crystal of rubidium chloride. Allow to stand for a short time and examine. Characteristic crystals of rubidium per- chlorate will separate, colored pink or violet through adsorption of the permanganate. The chlorate is only partially converted into the perchlorate, hence this test is not .always successful, and is of little value in complex mixtures. CHROMATES; BICHROMATES. a. Test with Silver Nitrate in nitric acid solution. See Silver, page 334; Chromium, page 357. b. Test with Strontium Acetate. See page 301. c. Bichromates give no separation of crystals with Manganous Sulphate; Chromates do. See Manganese, page 360. CYANIDES. a. Place the material in the gla.ss crucible of apparatus, Fig. 130, page 245; moisten with dilute sulphuric acid, cover with a slide bearing a drop of silver nitrate. If no tiny prismatic crystals are obtained and no clouding of the silver nitrate, cyanides are absent. If a clouding of the drop results, make a Iresh test, this time substituting for the sulphuric acid, a saturated MICROCHEMICAL REACTIONS OF THE COMMON ACIDS 377 solution of primary sodium carbonate; hydrocyanic acid will be set free and will give a characteristic silver cyanide. b. Set free the vapors of the acid and expose to them a drop of sodium picrate. A blood red solution results. CYANATES. a. To a drop of concentrated solution add at the center, a tiny crystal of cobalt acetate. The crystal will be immediately surrounded by a deep blue colored zone and a blue amorphous precipitate. The blue zone increases in diameter and eventually may reach the circumference of the drop. Upon evaporation deep blue tetragonal dendrites, and tabular and prismatic crys- tals of a compound corresponding to the formula K 2 Co(CNO)4 will appear. Note that to obtain this compound the cyanate must be in excess. With sulphocyanates tested thus a deep blue liquid is obtained on evaporation, but the blue dendrites which may separate have a different habit. Cyanides yield no blue, but a brown color instead. Even a small amount of cyanide will prevent the blue zone, but the crystal will be blue surrounded by a yellow or brown zone. b. Treat a drop with dilute sulphuric acid in the distilling apparatus, Fig. 130, page 245. Evaporate very gently almost to dryness; add a few fibers of freshly ignited asbestos and pro- ceed to test for ammonia. See Ammonium, Method A , page 286. With sulphuric acid cyanates yield carbon dioxide and ammo- nium sulphate. Precaution. Always make a blank test upon the reagents to be sure of their freedom from ammonium salts. FERRICYANIDES. a. Give off Vapors when heated with sulphuric acid which produce silver cyanide. See Cyanides, a, page 377. b. To the test drop add sodium acetate, then apply a solution of Benzidine Hydrochloride 1 by Method /, page 251. Light blue prisms and stars will soon appear. Ferrocyanides do not give this reaction. c. Give no color with dilute solutions of pure Ferric Salts. 1 Behrens, Z. anal. CJaem., 13, 432. 378 ELEMENTARY CHEMICAL MICROSCOPY FERROCYANIDES. a. Give a Blue Precipitate with Salts of Iron and a brown one with salts of copper in acetic acid solution. b. With Quinoline Hydrochloride yield upon warming cubical crystals. IODIDES. a. To a drop of solution add dilute sulphuric acid, a little potato starch and a tiny fragment of ammonium persulphate. The starch is turned blue or violet in the cold. See Bromides, page 375- b. The silver nitrate precipitate is insoluble in ammonium hydroxide; distinction from chloride and bromide. c. Yield characteristic hexagonal plates with lead nitrate. See Lead, page 323. IODATES. a. Dissolve in water, add a very tiny drop of dilute sulphuric acid, a little potato starch and finally a crystal fragment of morphine sulphate. Iodine is set free and the starch granules turn blue or violet. Iodides do not give this reaction; nor will iodates give reaction a under iodides. NITRATES. a. With Nitron Sulphate in Acetic Acid Solution. Apply the reagent by Method /, page 251. There is immediately formed a heavy precipitate, consisting of masses of exceedingly minute needles. In a few seconds sheaves of acicular prisms appear and later there are formed long thin prisms with square ends, giving polarization colors and parallel extinction. Nitron nitrate has a very low solubility even in warm water, hence the reaction is a delicate one. The sheaves of white crystals, appearing brownish by reflected light, are characteristic. In dilute solutions none of the salts of the common acids inter- MICROCHEMICAL REACTIONS OF THE COMMON ACIDS 379 fere save iodides and bichromates. With these salts there may be obtained crystals which closely resemble the nitrate but these crystals disappear upon even gentle warming; nitron nitrate will not. From concentrated solutions there may be obtained under favorable conditions, precipitates with chlorates, perchlorates, phosphates, chromates, bichromates, iodides, ferro- and ferri- cyanides, oxalates and tartrates, but in no case in dilute solu- tions with gentle warming should there be any difficulty in differentiating between such precipitates and the crystals ob- tained with nitrates. NITRITES. a. With Silver Nitrate there is obtained a felted mass of fine needles with long acicular prisms at the outer edge of the mass, changing into short stout prisms with imperfectly developed ends. These crystals are colorless under the microscope and do not show their greenish tint until viewed in masses by reflected light. b. With Potassium Iodide and Starch. Add to the drop to be tested a crystal of potassium iodide, then a little potato starch and finally a trace of dilute sulphuric acid. The hydroiodic acid set free by the acid is oxidized by the nitrous acid; iodine is liberated and stains the starch blue or violet or black. Always test the potassium iodide, with starch and dilute sul- phuric acid, to ascertain its purity and to be certain that no appreciable blueing of the starch takes place with the reagents alone. Only traces of iodine are liberated from iodide when treated with a crystal of morphine sulphate as described under iodates, page 378. OXALATES. a. With Strontium Acetate. See Calcium, page 291; Stron- tium, page 295. b. With Silver Nitrate or Lead Nitrate. See Calcium, page 291. 380 ELEMENTARY CHEMICAL MICROSCOPY PHOSPHATES. a. To the drop to be tested, add a drop of Nitric Acid. Then apply a drop of Ammonium Molybdate by Method /, page 251. Warm gently. Phosphates yield a yellow precipitate appearing amorphous under the microscope unless a magnification of over 200 is employed. A similar reaction will be obtained if silico- molybdates or arseno-molybdates are formed. This reaction is of value if arsenic and soluble silicates are absent and as indicating whether much or little phosphate is present. If a heavy precipitate is obtained, apply test b. b. To the Ammoniacal Solution add Ammonium Chloride and Magnesium Acetate, proceeding as described under Magnesium, page 305. Arsenates must be absent. Note. Phosphates frequently interfere with the detection of certain bases and must be removed before reliable reactions can be obtained; their removal may be accomplished by means of tin in acid solution. Acidify with nitric acid, add a few tiny bits of pure tin-foil and as soon as the reaction has ceased, heat to boiling. Cool and extract the material with dilute nitric acid. SILICATES. a. Treat the material upon a celluloid object slide with am- monium fluoride, sodium chloride and sulphuric acid. Sodium silicofluoride is formed. See Sodium, page 278. Boron, zirco- nium and titanium must be absent. SULPHATES. a. To the drop add a trace of Nitric Acid, then a drop of Cal- cium Acetate by Method /, page 251. Characteristic needles or prisms of calcium sulphate result. See Calcium, page 288. b. To the drop add a trace of Potassium Chromate, a trace of Nitric Acid and a drop of Silver Nitrate. Characteristic crystals of silver sulphate will be obtained, stained yellow through solid solution of the silver chroma te. See Silver, page 335. MICROCHEMICAL REACTIONS OF THE COMMON ACIDS 381 SULPHITES, THIOSULPHATES. a. To a drop of a solution of potassium iodate add a little potato starch and a small drop of dilute sulphuric acid. Ex- amine to see that no iodine has been set free. Add a fragment of the unknown. The starch is colored blue. b. To a moderately concentrated drop of copper sulphate apply a drop of a solution of the unknown by Method /// A, page 254. Warm gently sulphites, if pure and undecom- posed, yield at the most only a faint cloudiness thiosul- phates give a brown precipitate of copper sulphide and around the circumference of the drop lemon-yellow crystals of copper thiosulphate. SULPHIDES. a. The Silver Nitrate Precipitate was Black. b. Place a drop of solution or fragment of solid in the distilling apparatus, cover with a slide holding a tiny drop of silver nitrate and one of lead acetate side by side. Raise the cover and care- fully run in a drop or two of dilute hydrochloric acid. Cover quickly and allow to stand. Both drops turn black. c. Proceed exactly as in b but invert over the crucible a slide carrying a drop of sodium nitroprusside. A beautiful purple color results. SULPHOCYANATES. a. Give a Blood-red Color with dilute Ferric Chloride. b. Add Mercuric Chloride and Zinc Sulphate. There will be obtained the double sulphocyanate of mercury and zinc. See Zinc, page 308; Copper, page 340. Add a trace of copper and increase the delicacy of the reaction. TARTRATES. Note. Before testing for tartrates always neutralize any free mineral acid present. a. By means of Calcium Acetate. The solution may be neutral or acidified with acetic acid. 382 ELEMENTARY CHEMICAL MICROSCOPY Large, colorless, well-formed, highly refractive crystals are obtained. The solution to be tested must be concentrated, otherwise the calcium tartrate will not separate save on long standing. Ex- posure to alcohol vapors (Method VI, page 257) will hasten the formation of a crystal deposit. Magnesium salts greatly retard the separation of crystals of calcium tartrate. b. With Potassium Salts, tartrates yield characteristic color- less, highly refractive, orthorhombic, short, stout prisms of the primary salt KH^ILtOe. c. With Silver Nitrate. A granular precipitate only is obtained unless in very dilute solution, then there will be obtained tiny squares and rectangles and short, stout prisms giving a six-sided outline. Most other acids interfere with the detection of tartrates by means of the silver salt. CHAPTER XV. PREPARING OPAQUE OBJECTS FOR THE MICROSCOPIC STUDY OF INTERNAL STRUCTURE. In order that alloys and many other similarly constituted materials may be properly studied and their internal struc- tures ascertained it is usually essential that large or small pieces be ground down to a plane surface which may be so placed under the microscope as to lie at right angles to the optic axis of the instrument. It is further necessary that this plane sur- face shall be so smooth as to show no striations due to grinding, otherwise these parallel or irregular streaks will confuse the observer. Removal of the streaks is accomplished by polishing or, in other words, grinding with an abrasive so fine that the scratches made are so close together and so shallow that they will not be resolved by the objectives used in the microscopic examination. If these polished specimens are subjected to the action of various solvents, it will be found that in non-homo- geneous materials, certain components are easily dissolved and certain others are resistant. The specimen thus treated, is said to have been etched, and when the etched surface is examined a more or less marked crystalline structure is visible. Through the judicious selection of the proper etching liquids we are able to bring into view different components or phases and thus trace the changes in structure through changes in percentage compo- sition, or through changes in the temperatures to which the specimens have been submitted. Or instead of submitting the polished surface to the action of a corrosive liquid, we can rub it upon a thick, soft cloth charged with a fine abrasive powder. The softer components will thus be more rapidly worn away than the harder; again we obtain evidence of a more or less marked crystalline structure. The specimen is no longer spoken of as having been etched, but 383 384 ELEMENTARY CHEMICAL MICROSCOPY is said to have been polished in relief. Since in almost all the materials commonly studied we deal with components differing in hardness, it is exceedingly difficult to obtain polished speci- mens which do not exhibit some relief polishing. Practice and a light touch are the only effective preventives. The wearing or cutting off of irregularities so as to obtain a flat surface is termed roughing. Roughing is most easily accom- plished by holding the specimens against rapidly revolving ab- rasive wheels. The most useful American abrasive wheels are emery, co- rundum, alundum, crystalon and carborundum. Emery and corundum are natural products, while alundum, crystalon and carborundum are products of the electric furnace ; the first three mentioned consist of crystallized alumina, the last two consist of crystalline carbide of silicon. Of these, emery cuts or wears away specimens the least rapidly, crystalon and carborundum the most rapidly. All three steps, grinding, polishing and etching, require patience, practice and a certain inherent technical skill. Prac- tice, and practice alone, will enable the student to properly prepare specimens. The selection of the proper sequence of abrasives, the right pressure of the specimen against the grind- ing material, the rate of speed or motion in grinding and polish- ing all enter into the preparation of the specimen. No specific directions can, therefore, be given, but merely a general outline of the steps to be taken and the special precautions to be observed. So, too, in the etching much depends upon the individual. The proper concentration of reagent (which differs for different alloys of the same type) , the way in which the specimen is immersed or submitted to the action of the reagent, the time of exposure, temperature of the room and reagent, thoroughness of removal of the etching liquid by washing, etc., each enters largely into the preparation of really satisfactory specimens and all con- tribute to the elucidation of the problem or to the confusion of the investigator. Grinding wheels are made from powdered abrasive mixed with a suitable binder, pressed into moulds and fired in an oven. PREPARING OPAQUE OBJECTS 385 The character of the binder and the degree of incipient fusion characterizes a wheel as hard or soft. The degree of hardness or softness is technically spoken of as the grade or hardness of the wheel. American manufacturers usually indicate the grades of their wheels by letters of the alphabet, but the scale of hardness as indicated by the letters is by no means uniform with different manufacturers. Consequently, a letter indicating a grade can- not be interpreted without reference to the scale of hardness of the particular firm from whom the wheel was obtained. For example, we find that a wheel marked U may be "hard" as supplied by one firm, but if we purchase a U grade from another firm we will obtain a "very soft" wheel. In selecting wheels for grinding specimens, it is safe to be guided by the general rule that a soft wheel will cut more rapidly and deeper than a hard one, will clear itself more readily, but is more easily worn away, and therefore more liable to be spoiled. The soft wheels as a rule must be run at higher speeds. Hard wheels on the other hand tend to glaze over, cause more heating of the specimen and often yield aggravated cases of surface films or surface flow of soft components, but they cut slower, hence do not so deeply score or furrow the specimen through injudicious pressure and may be employed to better advantage when only low speeds are available. Besides the grade or hardness of grinding wheels as influencing their suitability for certain work, the diameter and the uniform- ity of the individual particles employed in building up a wheel must be taken into account. The size of the component par- ticles is called the grain or grit. Grain is obtained in manu- facturing by screening the abrasive powder. The number of linear meshes to the inch through which the powder will pass is the grain number of the wheel. For example, in a wheel marked 50, the component particles will pass through a sieve having fifty meshes to the inch. The grain numbers employed by different manufacturers are not comparable because the size of wire employed in the sieves used for the grading is not always the same. Since it is the number of linear meshes to the inch and not the diameter of 386 ELEMENTARY CHEMICAL MICROSCOPY the opening that is recorded, the size of the wire greatly influ- ences the screened product. Although for industrial purposes abrasive wheels may be said to conform closely to the grade and grain indicated by the manufacturer, it will be found that in preparing specimens for microscopic study, wheels are not easily duplicated and if we purchase a wheel to replace one accidentally ruined we are apt to find that it will not do just the work of the one lost. Wheels of softer grade and coarser grain (at high speeds) can be used for roughing chilled iron and steels, hard and of high tensile strength, than for material like brass soft and of low tensile strength. No single type of wheel as to grade and grain will answer for all purposes. A laboratory in which a great variety of work is to be done will therefore require a series of wheels. TABLE VI. CHARACTER OF ABRASIVE WHEEL REQUIRED. Grain or grit. Grade or Hardness. Alloy, aluminum type Alloy, brass type 20 to 36 20 to 46 Hard Hard Alloy, bronze type 20 to 36 Hard Alloy, nickel type 20 to 36 Medium-hard Iron, cast 3O to g 1! II 1 < 0) "S a J"" II 6 a ' Is * ss> 'eg? || I 1 ex S c o 3JI 8ii HH ^ g PH O M d 'I " c 3 || ^ a a P3 o Sfe 1 1 3 o B 1 g B 1 '? 1 1 o. $ ll ^C/2 ^ o S go 3s, >> M 1| || Ii O ^> || M " -So d II 2 u 00 OO s I P& o. * |S g Jell, Vanadium V=5i.o g 4 Columbium Cb=93.s Antimony Sb = I2O.2 I 1 Tantalum Ta= 181.5 I S 1 d M S "^ d o i tf M > 82 g 3 pi B H ^ c? ' L? O * P. <5 Ju d J 4 s^ doo" 'B J |g 3 b " If | O ^ jjjjT OO c3 bO OO u CS3 S ii MM ctf'd 00 ii pqpq ! Is g? ss? go ^o Bi$ -. OO CUOO 5 1 2 ^ 0. 3 s s 1? If .2 ll ^ M 'S g; rt fj 5 If If Is P u o K 53 IM O fS bo 06 1 s 1 p. B" go" S g a bo 2 JT ^ ll ^ tf o M (L) >> u d II 1 "a) 3^9 399 400 INDEX PAGE Analyzer 162 Antimonates 354 Antimony, common salts 351, 352 detection with cesium chloride 352 Apochromatic objectives 3 Applying reagents, methods 250-272 Arc lamps, microscopic 133, 134 Areas, measurements of 216 Arsenates, detection of 351, 374 Arsenites, detection of 351, 374 Arsenic, common salts 348 detection of 349 as arsine 349 by reduction 350 Artificial daylight 132 Artificially colored crystals 170 Atomic weights 396 Axial light, test for 35 Ball-and-socket stage 117 Barger method for molecular weight determination 216 Barium, common salts 295, 296 detection by oxalic acid 298 potassium bichromate 301 ferrocyanide 300 sodium bicarbonate 303 sulphuric acid 296 Barnes pipette, bottles with 121 Bead tests 268 Biaxial crystals 171 refractive indices of 194 Bichromates, detection of 376 Biltz cell 60 Bisectrix 173 Bismuth, common salts 354 detection by cesium chloride 355 oxalic acid 356 potassium bichromate 356 sulphate 355 water 355 Blast lamps, small 128 Blue glass with Abbe condenser 37, 132 Books, reference 397, 398 Borates, detection of 375 Brinell hardness testing method 147 Brittle material, grinding and polishing 390 Bromides, detection of 375 INDEX 401 PAGE Brownian movement 55 Burners, micro- 127 Cadmium, common salts 316 detection by ammonium mercuric sulphocyanate 317 oxalic acid 318 sodium nitroprusside 318 Calcium, common salts 287, 288 detection by oxalic acid 291 potassium f errocyanide 300 sodium bicarbonate 303 sulphuric acid 288 Camera lucida 102 Carbonates, detection of 375 Cardioid dark-ground illuminator 67 ultramicroscope 66, 67 Casseroles 248 Cations, testing for 367 Centering Abbe condensers 34 objectives 47, 60 stage of microscope 164 Centrifuge 233 tubes for 234-236 Cesium chloride as reagent 368 double chlorides 348 Celluloid object slides 125 Chlorates, detection of 376 Chlorides, detection of 376 Choice of abrasive wheels 386 Chromates, detection of 376 Chromium, common salts 356 detection by color of salts 356 cesium sulphate 356 in alloys 358 Circular polarization 160 Cobalt, common salts 365 cyanate 366 detection by potassium nitrite 366 sodium phosphate 367 ammonium mercuric sulphocyanate 366 Colors with Abbe condenser 37, 132 Coma 2 Compensating oculars 13 Compound microscope, optics of I Condensers, Abbe 32, 33, 34 Jentzsch ultra- 71 numerical apertures of .... 32 402 INDEX PAGE Condensers, reflecting 45 Contrast micrometers 154 Converging polarized light 173 Coordinate ruled cells 210 ocular micrometers 208 object slides 210 Copper, common salts 339 detection by ammonium mercuric sulphocyanate 339 cesium chloride 341 . potassium ferrocyanide 341 triple nitrite reaction 340 Cotton and Mouton ultramicroscope 69 Counting cells 210, 211 Cover glass, correction for . -. 3 Cross-hairs, testing 163 Crossed nicols, testing for 161, 163 Crystal angles, measurement of 177 Crystals, axes of 168 faces of 168 symmetry of 168 for determination of refractive index 203 Crystallographic concepts 167 Cubic system, characteristics of 180 Cups, platinum 248 Cyanates, detection of 377 Cyanides, detection of 376 Dark-ground illumination 40-50 illuminators, adjustable 43 adjustment of 46 numerical aperture in 43 path of light rays in 41, 42, 45 thickness of object slides for 49 Decantation 230 washing precipitates by 233 Dendrites 169 Differential color illumination 51 Dimorphous crystals 169 Directions of elasticity 172 vibration 172 Disk vertical illuminators 77 Dispersive power of liquids 190 Distillation 244 apparatus 245 Distilling tubes 247, 248 Drawing cameras 102 adjustment of 104 INDEX 403 PAGE Drawing oculars 105 Draw tube 4 Ebonite tubes 121 Elasticity, axes of 172 Electrochemical series 253 Estimation of molecular weights 216 Etching 391 liquids 3Q2-394 Evaporators for microchemistry 126 Experiments in crystallization 182, 183 Extinction 171 angles 179 Extraordinary ray 161 Eyepieces, compensating 13 cross-haired 163 drawing 105 goniometer 14 Huygens n micrometer 149, 153-155 negative 1 1 net ruled ' 208 positive ii projection 14 Ramsden 1 1 Eye-point 12 Ferricyanides, detection of 377 Ferrocyanides, detection of 378 Fibers, textile, use of, in analysis 260 preparation of, as reagents 271 Files, method of using, to prepare metals for study 388 Films of material for analysis, preparation 255 Filter tubes 237, 238, 239 Filtration 236 Fluorescence microscope 51 Fluorides as reagents, precautions 268 Forceps 122, 123 Fusions 248 Gas lamps 26 Gases, testing for evolution of 263 Glass, refractive index of 200 Glass rods 121 Glycerin, use of 254 Grade of abrasive wheels 385, 386 Grain of abrasive wheels 385, 386 404 INDEX PAGE Graduated circles, testing of 165 Grinding material for analysis 249 opaque objects for microscopic study 383-391 wheels 384, 385, 386 Ground glass with Abbe condenser 36 loss of light in using 136 Habit of crystals 1 70 Half-shadow illumination 189 Hemihedral crystals 169 Hemimorphic crystals 169 Hemispheres for orientation 115, 116, 117 Hexagonal system, characteristics of 181 Holohedral crystals 169 Hot stages 222, 223, 224 Hydrofluoric acid as a reagent 121, 268 Idiomorphic crystals 169 Ignition 248 Illumination 3~53 dark-ground 40 differential color 51 orthogonal 50, 61 reflected light 37~39 transmitted light 30, 31 ultraviolet ray 51 Illuminators, vertical 76 Immersion, method of refractive index determination 184-196 method, liquids for 201, 202 objectives 5 ultramicroscope 72 Interfacial angles 168 Interference colors 174 figures 173 Interpretation of appearances with transmitted light 31 reflected light 38 lodates, detection of 378 Iodides, detection of 378 Iron, common salts 362 detection by oxalic acid and barium salts 299 potassium ferrocyanide 362 sulphocyanates 310 Isometric system, characteristics of 180 Isotropic bodies 159 crystals 171 refractive index of 185 Kryptokinetic motion 55 INDEX 405 PACK Lead, common salts 323 detection by hydrochloric acid 325 metallic zinc 329 potassium iodide 323 triple nitrite reaction 327 Leitz metallurgical microscopes 91, 100 Lens holders 1 18 paper 9 Light grasping power of objectives 5, & Liquids for determination of refractive index 201, 202 determination of refractive index of 196 Litmus-silk fibers 271 Luminescence microscope 51 Magnesium, common salts 304 detection by sodium phosphate 305 uranyl acetate 275 Magnification, limit of 15 determination of 148 Manganese, common salts 359 detection by chromates 360 fusion 361 sodium bismuthate 361 sodium phosphate 361 oxalate 359 Mazda lamps for microscopy 136 Measurement of areas 216 Measurements, microscopic 142 of thickness 200 volumes 216 Mechanical stages 113 testing graduations 114 Melting points, determination of 220 table of 395 Mercury, common salts 318 detection by ammonium sulphocyanate 322 potassium iodide 321 sublimation -. . . 319 determination of 214 Mercuric and mercurous compounds, differentiating 320 Metallurgical microscopes 90 Metals, grinding, polishing and etching 383 Microburners 127 Microchemical methods 228, 250 Micrometers, contrast 154 filar 154, 155 step 153 406 INDEX PAGE Micrometry 142 by means of camera lucida 147 fine adjustment 157 ocular micrometer 149 projected scale from Abbe condenser 155 Micrometer scales, adjustment of 151, 152, 153 Micrometric microscopes 144, 145, 146 Microscopes, chemical 17, 19 specifications for 17, 18, 19 compound, optics of i comparison 23, 26, 27 fluorescence 51 hot stage 29 large stage 22 metallurgical 90 as polarimeters 140 Microspectroscopes 106 adjustment and calibration no, in, 112 Micellae 56 Monoclinic system, characteristics of 181 Mortars, agate , 249 Mounting opaque objects for study 89 Nernst lamps for microscopy 135 Neutralization 254 Nickel, common salts 363 detection by glyoxime 363 phosphate 365 triple nitrite reaction 365 Nicol prism 160 Nitrates, detection of 378 Nitrites, detection of 379 Nosepieces 137 Object slides 123 for use with fluorides 125 Objective changers 137 Objectives, achromatic 2,3 adjustable 3 angular aperture of 4 aplanatic 2 care of 9 designation of i for dark-ground illumination 44 function of i illuminating power of 8 immersion 5 INDEX 407 PAGE Objectives, light grasping power of 5,8 numerical aperture of $ penetrating power of 7, 8 photographic 9 selection of 8 variable 6 vertical illuminator 79 Oblique extinction 171 illumination with Abbe condenser 31, 33 in refractive index determinations 189 Ocular micrometer ratio 151 Oculars, care of 14 comparison 24 compensating 13 cross-haired 163 goniometer 14 Huygens 1 1 micrometer 149, 153-155 negative 1 1 net ruled 208 par focal 13 projection 14 positive ii Ramsden 1 1 Oil globules, optic behavior of 187 Optic axes of crystals 171 Optics of compound microscope i Ordinary ray 160 Orientating devices 115 Orthogonal illumination 5 Orthorhombic system, characteristics of 181 Oxalate, potassium, as reagent 368 Oxalates, detection of 379 Parallel extinction I7 1 Paraboloid dark-ground illumination 47 Penetrating power of objectives 7, 8 Periodic system of Mendelejeff 396 Phosphates, detection of 380 Platinum cups 248 wires 121 Pleochroism 176 Polarizer 162 Polarization colors 1 74 by reflection 166 tube 139 Polarized light J 59 408 INDEX PAGE Polishing . 384 Polymorphous crystals 169 Potassium, common salts 211 detection by chloroplatinic acid 281 perchloric acid 284 Prism vertical illuminators 77 Pseudomorphs 169 Quantitative microscopic analysis 205, 214, 215 Quartz object slides 124 Quartz, refractive index of 200 Radiants for microscopic illumination 132-137 Reagent cases 1 20 containers 119, 121 Reagents, methods of applying 250-272 Reference books 397, 398 Reflected light illumination 37 Refractive index, determination of 184 liquids for 196 calculation of, for liquid mixtures 191 biaxial crystals 194 uniaxial crystals 194 of typical crystals 204 Reichert metallurgical microscope 94 Resolving power of objectives 6, 7, 8 with dark-ground illuminators 44 Rotating apparatus 115 Sedimentation glasses .< 88, 236 Sedgwick-Raf ter counting cell 211 Selenite plate 174 Shop microscopes 99 Silver, common salts 330 detection by arsenic acid 337 bichromates 334 hydrochloric acid 331 Silicates, detection of 380 Skeleton crystals 1 70 Slit ultramicroscope 57 path of rays in cell of 61 Sodium, common salts 273 detection by bismuth sulphate 276 silicofluorides 278 uranyl acetate 274 Soft metals, preparing for study 390 Solubility, testing for 228 Spatulas for chemical microscopy 122 INDEX 409 PAGE Sphere-crystals -. 169 Spherulites 169 Spectroscopic ocular 106 Stage, attachable mechanical 113 method of centering 164 Step micrometer 153 Strontium, common salts 292, 293 detection by bichromates 302 oxalic acid 295 sodium bicarbonate 303 sulphuric acid 293 Sublimation 240 Subliming cell 243 point determinations 220 Sulphates, detection of 380 Sulphites, detection of 381 Sulphocyanates, detection of 381 Sulphuric acid, reagent in qualitative analysis 367 Supports for objects 123 Symmetrical extinction 171 Tables for microscopic work 129 Tartrates, detection of 381 Testing graduated stages 165 polarizers and analyzers 165 Tetragonal system, characteristics of 181 Thickness, measurement of 200 Thomae cell 60 Thiosulphates, detection of 381 Tin, common salts 346 detection by cesium chloride 347 Tongs 129 Trichites 169 Triclinic system, characteristics of 182 Trimorphous crystals 169 Tungsten lamps for microscopy 136 Turmeric-linen or -silk fibers 271 Ultracondenser, Jentzsch 71 reflecting 65 Ultramicrons 56 Ultramicroscopes 54, 57, 66, 67, 69, 72 Ultramicroscopic studies, preparing solids for 63 Ultraviolet ray illumination . . .' 51 Uniaxial crystals 171, 194 Vapors, tests involving 266 Velocity of abrasive wheels 387 410 INDEX PAGE Vertical illumination, appearances in 38, 39, 80 Vertical illuminators 76 adjustment of 78, 88 auxiliary stage with 88 disk 77 lamp for use with 87 Leitz 82 Nachet 81 objectives for use with 79 polarized light with 81 prism 77 stand for use with 84 Tassin 84 Watch glasses 126 Weight, determination of 213 Work tables 129-131 Working distance 2 Works microscopes 99 Zinc, common salts 307 detection by ammonium mercuric sulphocyanate 307 oxalic acid 311 sodium bicarbonate 311 nitroprusside 315 sulphide fibers, preparation of 272 as reagent 368 I UNIVERSITY OF CALIFORNIA LIBRARY BERKELEY Return to desk from which borrowed. This book is DUE on the last date stamped below. NOV 23 1948 NOV 1 9 1949 / 171 ** * *' ^ * . 72- LD 21-100w-9,'47(A5702sl6)476 309481 Engineering Library UNIVERSITY OF CALIFORNIA LIBRARY