r$ai)ii$loo SECOND EDITION I9O5 Textiles TS14-93 D7 1^05 PRESEN 1 the draper company HOPEDALE, MASS. She 1. 3ft. Mill library North (taroltna £>tatp Unitierattu, Textiles TS]>93 D7 1905 RARE BOOKCASL RARE BOOKCASE THIS BOOK IS DUE ON THE DATE INDICATED BELOW AND IS SUB- JECT TO AN OVERDUE FINE AS POSTED AT THE CIRCULATION DESK. 100M/7-85 THE FIRST NORTHROP LOOM. Designed for the weaving of Print Cloth and Sheetings solely. Used with great success on plain two-harness weaves by our original customers. It was this model that first proved a weaver's capacity to run sixteen looms. It incorporated the inventions of: — James H. Northrop, Charles F. Roper, William F. Draper, George Otis Draper, Edward S. Stimpson and [ohn W. Keki.kv. The loom frame and other conventional parts were designed for the Hopedale Machine Company under supervision of Oren B. Smith. The II. M. Co. was incorporated with the present Draper Company in 1896. LABOR-SAVING LOOMS. (SECOND EDITION.) A BRIEF TREATISE ON Is?lain Leaving AND THE RECENT IMPROVEMENTS IN THAT LINE WITH SPECIAL REEERENCE TO THE . . . N0HT8R0P iioojws MANUFACTURED BY DRAPER COMPANY, HOPEDALE, MASS., u. s. ^. 190 5 COPYRIGHT 1905, BY DRAPER COMPANY. WRITTEN AND COMPILED BY GEORGE OTIS DRAPER, SECRETARY OF THE DRAPER COMPANY. PRINTED BY Cook & Sons, Milfokd, Mass. PREFACE. This book cannot serve as a detailed catalogue by which the purchaser can always note the exact nature of the device- we shall continue to sell, as improvements are often unexpectedly invented. We can hardly expect to publish a work of this size at short intervals, but shall try to keep it reasonably up to date by amended additions. New matter will be inserted in the final pages of each new edition. OUR HOPEDALE PLANT IN 1904. Scale. 315 feet to the inch. About 2j acres of Horn- space in all. FORMER LITERATURE ON THE NORTHROP LOOM. 1895. Circular — The Advent of the Northrop Loom, issued April, 1895. Essay, The Present Development of the Northrop Loom, delivered by George Otis Draper at the meeting of the N. E. Cotton Manufacturers' Association at Atlanta, Ga., Oct. 24, 1895. Printed in Vol. 59 of the Transactions. 1S96. Papers on The Northrop Loom, by F. M. Messenger, John H. Ilines, H. D. Wheat, and discussion by Wm. F. Draper. Arthur II . Lowe, George F. Whittam and W. J. Kent, April 29, 1S96, printed in Vol. 60 of the Transactions of the X. E. Cotton Manufacturers' Association. Chapter in Facts and Figures, on the Northrop Loom, pub- lished by George Draper iV; Sons in the spring of 1896. Speech of Hon. Wm. C. Lovering, published in the Scien- tific American of May 2, 1S96, and other papers, containing pertinent reference to the loom. Pamphlet— The Looms of the South, by F. B. de Berard, issued March, 1896, containing detail of savings from use of the Northrop Loom in Southern mills. Speech of I Ion. Charles Warren Lippitt, published in the Manufacturers' Record of June 19. and papers generally through- out the country, giving the history of the Northrop loom devel- opment as illustrative of the educational influence of manufac- turing. '897- Pamphlet — Instructions for Running Northrop Looms, issued by George Draper & Sons, January, 1897. Pamphlet — Instructions Pour la ConJuitc de Metiers Northrop, issued by the Draper Company, 1S97. Circular — Our Common Loom, issued by the Draper Com- pany. June. 1897. Circular — The Triumph of the Northrop Loom. November, 1897. 1S9S. Circular — Our Connection with the Art of Weaving, issued by the Draper Company, April. 189S. Circular — Take-up Mechanism, issued by the Draper Corn- pan;'. 189S. Article — Industrial Investigations, by Jacob Schoenhof, in The Forum for October, 189S. Referred to the great savings of the "Automatic loom.'' as affecting differences in cost of production. 1S99. Pamphlet — Instructions for Running Northrop Looms, ( Re- vised Edition ) issued by the Draper Company. January, 1S99. Pamphlet — Machinery and Labor Displacement, by George Gunton, issued by the Gunton Institute, containing pertinent reference to the Northrop Loom as a labor-saving invention. 1900. Circular — The Advance of the Northrop Loom, January , 1900. Pamphlet — Factory Conditions in the South, January 20, 1900. by George Gunton, in Gunton's Lecture Bureau course. Paper on Method of Cost Finding, by Wm. G. Nichols, delivered at a meeting <>t" the X. E. Cotton Manufacturers' Asso- ciation at Boston. April 26, 1900. Printed in Vol. 68 of the Transactions. Essay on Improvements in American Cotton Machinery, by George Otis Draper, delivered at a meeting of the Southern Cotton Spinners' Association at Charlotte, X. C, May 18, 1900. Printed in the Association records and various periodicals. 1 90 1 . Chapter in Textile Texts, published by the Draper Com- pany, spring of 1 90 1. Various articles in publication, Cotton Chats, started in July, 1901, and continued since. Circular on Important Discovery, relating to method of spin- ning to prevent bunches in cloth, August, 1901. 1902. Circular on The Keene Drawing-in Machine, January, 1902. References in Census Bulletin Xo. 21 ^. June 2S, 1902. 1903. Circular on The Northrop Loom, issued by the British Xorthrop Loom Co.. January, 1903. Essay on Continued Development of the Northrop Loom, deliv- ered by General Draper at a meeting of the X. E. Cotton Man- ufacturers' Association in Boston. April 23, 1903. printed in Vol. 74 of the Transactions. Various references in a book, V'/ie American Ct>ti<>// Industry, by T. M. Young, published by Charles Scribners' Sons, 1903. IO Chapter on Northrop Loom in Textile Texts, second edi- tion, issued December, 1903. Essa) on The Development of the Northrop Loom, delivered before the Providence Society of Mechanical Engineers by George Otis Draper, printed in Providence Journal. Dec. 28, 1903. and other trade journals. 1904. Circular on List of Northrop Looms Sold, issued January, 1904. Article on Evolution of the Cotton Industry, published in Gunton's Magazine for February, 1904. Pamphlet — Labor Saving Looms, (First Edition ) . Article on Is there a limit to rising wages? published in Gun- ton's Magazine for July, 1904. Reference in article on The Cotton Industry in the Cosmo- politan Magazine for July, 1904. Address of President Herbert E. Walmsley to the New England Cotton Manufacturer-' Association, published in Vol. 77 of the Transactions. Article on The Northrop Loom in England, by H. P. Gregg, published in British papers Dec. 31. 1905. Varied press comment on labor conditions as affected by automatic weaving. References in Cotton Manufactures in Massachusetts and the Southern States, published by the Massachusetts Bureau of Sta- tistic- of Labor. 190^. ( List complete only to July i-t. ) COLLECTED EVIDENCE. Also Northrop Loom History, Vol. I, 1889-1892 — 574 pages. Northrop Loom History, Vol. II, 1893-1896 — 1097 pages. Northrop Loom History, Vol. Ill, 1897-1900 — 818 pages. These books are by the Secretary of the Draper Company and were compiled for general reference and use by counsel during litigation. They contain the history of the experiments and development of the loom, and associated matters of inter- est. Their contents are naturally priyate, and not intended for general circulation, although the public is therefore depriyed of an acquaintance with a unique mechanical romance. It is believed that no other volumes of like size were ever prepared for such a purpose. In our circular of November, 1S97, we had a word to say to possible competitors, which still seems pertinent. We there- fore reprint a portion as a few unfortunate experimenters failed to note its truth on first appearance : "There are doubtless many bright men who will in the next tew years give time and toil in the endeavor to evade the claims of our pat- ents while producing similar mechanism. In view of the many other fields for inventive skill we ask — Is it worth the while? We are un- doubtedly ili«' first in the field and legitimately entil led to a fair reward for the expenditure of money, loss of time and consumption of brain energy. Our success is no vagary of chance or lucky stroke of for- tune. Every step in advance has hern gained after constant thought and experiment, with ten failures tor every Success. The patent othce has recognized the novelty of our device- by broad basic claims. We have searched the records here and abroad, and have proof that we are pioneers in our line. We shall defend our rights in the courts with the obstinacy of conviction, if such methods are necessary. We bave no wish for chance to show our strength. A lawsuit involves a waste ol energy for one side at least, and an expense Cor both. We appreciate these facts after thirty vears of continuous litigation. 1 " I 2 THE ART OF WEAVING. The process of weaving cloth consists in interlacing a con- tinuous thread amidst a series of parallel threads. Without giving an exhaustive history of the art it may be pertinent for further comparison to note down certain steps in its progress. It is fairly well established to-day that woven goods were used as clothing by the ancient Egyptians fully 6000 years ago. I have seen in Switzerland a preserved section of a net woven of twisted threads supposed to have been the work of the Lake dwellers in pre-historic times. In the ear- lier processes it is probable that the warp threads were stretched on pegs, the weft bring inserted by the fingers. In such weaving the warp threads usually lie vertically and in fact this is the method used l 3 to-day in producing rugs in the Orient with short wefts. With the use of longer weft also came the use of a stick with a hooked end for pulling it into position. If we are to form our further comparisons on a plain print cloth of the present width of 28 inches containing 64 threads of warp and 64 of filling per inch, it is possible that the rate of weaving by this method on such goods could be figured as low as one pick per minute per opera- tive in the earliest use. Cloth is still woven by this method in India, although a harness motion is added. History gives no rec- ord of the time at which the warp threads were divided by har- nesses and the shuttle introduced. References are made to shuttles in the Bible and other ancient books. It is probable that the general styles of hand loom weaving were very similar for many centuries without definite change until the invention of the fly shuttle by John Kav in 1733. At this time, in weaving broad cloth it was necessary to have two weavers at least, one at each end of the lay to throw the shuttle to the other. By Kay's invention one of these two men was dispensed with and even on nar- row weaving a weaver could produce at least twice as much cloth J 4 per day. No literature that I have run across gives any figures of production on the looms of this period and considering their crudeness in other lines, it is perhaps fair to assume that they could not produce at a greater speed than 20 picks per minute before Kay's time, probably averaging less. Kay's invention caused great commotion amongst the weaving trade and he was forced by persecution to leave the country. Cartwright's power loom patent was granted in 1 7S5 . Authorities differ as to the success of his first looms, some claiming that the early use was of no importance, while others refer to a mill of 500 looms in which Cartwright was interested, as being destroyed in 1790 by a mob in sympathy with the hand loom weavers. Whatever the cause, there were as late as 1S13 but 2400 power looms in all Great Britain. The first power loom was introduced in Waltham in America in 181 S- At this period one operative was required to each loom, as they had no weft stop motion and no self acting temples, the weaver having to intermittently move the flat wooden pieces with points at the end which held the cloth extended at the selvage. The inven- tion of the rotary temple by Ira Draper in 1S16. as developed set eral years later, allowed the operative to tend two looms instead of one. The speed of the common power loom at this time does not seem to be recorded, but it was probably between So and 100 picks per minute. In 1820 it is figured that there were about 15.000 power looms in England and Scotland and in 1830 perhaps 60,000. Even as late as 1840 there were said to be 250.000 hand looms still running. At this time weavers in England were not given more than one loom each, although in America they were running two looms, as the English manufacturers did not adopt the rotary temple so early as our American manufacturers. As to the comparative production of the common looms at this period, it is difficult to find any accurate basis of comparison. Hand looms were weav- ing print cloth as late as 1896 in Bohemia, where the production figured on 64 picks per inch in the cloth at ten hours per day would give an average of 35 picks per minute. I have been given figures of hand loom production recently that would sug- gest a possible speed of 60 picks per minute. About 1840 the weft fork began to he introduced and in America, by 1850, print looms were running at a speed of 150 picks per minute, with one operative tending four looms. Perhaps they even ran faster in England, hut the operatives only tended two looms. From this period to 1S95 the plain loom was not materially changed in principle, and vet the perfection of detail had brought the speed of the American plain loom up to 190 picks with one good weaver tending eight looms, while the English operative with looms at a speed of 220 picks per minute was tending four looms, though usually with a helper. In 1895 the Northrop looms then introduced immediately allowed one weaver to run 16 print looms at 190 picks and to-day it is assumed that a good weaver with the Northrop loom on prints can easily tend 24. In calling the speed of the American print loom 190 picks it is not intended to give a maximum. American print looms have run over 200 picks, hut such is not the general practice. In the same way English looms have run higher than 220 picks, hut the figures given are assumed as fair for the purpose of com- parison and as illustrating the general practice. Arranging a table of comparison, if we take 24 Northrop looms at 190 picks per minute, we have a total of 4560 picks. On the same basis, without allowance for stops, eight common looms would show 1520 picks, or four English common looms at 220 picks, 880 picks per weaver. The perfected hand loom would show perhaps 30 to 60. The power loom of 1S50. at 150 picks, with four to the operative, would show 600 picks per minute, while the loom of about 1840, before the weft fork, would show with two looms per weaver at perhaps 130 picks 1 6 per minute, 260 picks. Before the temple, the loom at 100 picks with one loom per weaver, would give 100 picks, while the hand loom before Kay at 20 picks, the loom of the middle ages with a possible 10 picks, and the loom of ancient history with a possible one pick per minute, brings our table down to a concrete illustration, which, even if faulty in detail, allows a comprehensive idea of the wonderful advantages since the earli- est application of the art. The Northrop loom in eight years has added over 3000 picks per minute per operative : the devel- opment since 1850, 920 picks; the inventions from 1830 to iS^o, 370 picks: the inventions from 1S20 to 1S30, 130 picks; the inventions from Kay to 1S20 would add 80 picks, the progress previous to this time being represented by 20 picks. It will thus be seen that within two centuries the productive power of the operative has been increased 22S times, and it is also seen that the advantages of the Northrop loom show twice as much in product as all of the other inventions put together. Historv is practically silent as to the inventors who supplied the earlier devices employed on the hand loom. It is not, there- fore, known who suggested the idea of the harness motion with its shifting heddles, the swinging lay with its reed, the take-up roll, the early jaw temple, and the shuttle itself. Starting with Kay, the development before the Northrop loom is shown by the following table, material for which is collected from stand- ard works on weaving. No attempt is made to include the vari- ous inventors of fancy loom devices, including the jacquard motion, the dobbv motion, and other ingenious developments. It might be well, however, to note that the earliest mechanism for fancy weaving; namely, the drop box. was invented by Robert Kay. son of John, in 1760. In preparing the table it has also been thought well to limit the inventions to show only the anticipation of the general principles employed. It is '7 impossible to properly note any l>ut the pioneer inventors, and the dates given are usuall) those of their patents. Verj possi- bly more credit is due other inventors not mentioned, for their perfection of ideas that otherwise would not have been useful. 17^. Fly shuttle. John Kay. 1786. Power loom. Edmund Cartwright. (First suggestion of warp-stop-motion, weft-stop-motion, positive let-off and take-up.) 1796. Over-pick, hinder, protector, and frog, Richard ( rorton. 1796. Ratchet take-up. Rohert Miller. [803. Shedding motion, John Todd. 1S10. Revolving temple, Ira Draper. [821. Multiple harness motion. Robert Bowman. 1828. Complete power loom with modern over-pick, William I Hckinson. 1830. Complete power loom, Richard Roberts. (These two instances of complete power looms are men- tioned as showing a general development oi ideas not noted in detail, which together produced practical weaving machines. ) [831. Weft fork, claimed by Clinton G. Gilroy. 1834. First shuttle-changer, John Patterson Reid and Thomas Johnson. [834. Weft fork, claimed by Ramsbottom and Holt. 1S38. Picker check. Rohert Pickles. 1S40. Improved temple. George Draper. 1841. Weft fork improvements, William Kenworth) and lames Bullough. 1S42. Loose reed. James Bullough. 1545. Loom brake, John Sellers. 1546. Parallel shuttle-motion for under-pick loom. War- ren W. Dutcher. tS [851. Reciprocating temple, Elihu and Warren \\ . Dutcher. 1857. Automatic let-off, Snell and Bartlett. [859. Rocker motion. W. Stearns. 1863. Loose frog, George Draper. 1867. Double beam let-off, Cottrell and Draper (George). 186S. Practical self -threading shuttle, J. A. Metcalf. r868. Broad loom shuttle-motion. J. Lvall. 1S69. Inside catch shuttle. J. II. Cohnrn. There is quite a lapse between 1870 and 1890 in which no vcrv important patents on plain looms were granted. In fact. looms made before 18^0 continued running for years in compe- tition with those built long after, the more modern looms not showing any notable advantage, except perhaps in heavier con- struction and higher possible speed. It must be remembered that I am still referring to the plain, common loom, not in any way intending to disparage the remarkable advance in the range of fancy loom devices in that period, including the hair-cloth loom, pile fabric loom, tape loom, etc.. etc., etc. Owing to an error in the index of the official British publi- cation of Abridgements of the .Specifications relating to Weav- ing, it was only recently that we discovered the first patent in which the idea of changing shuttles automatically is referred to. Such a reference occurs in that granted John Patterson Reid and Thomas Johnson. No. 6^79, in the British Patent Office, dated March 20, 1834. The specification refers to a number of different inventions, contemplating the weaving ol four webs of cloth at once in a vertical loom. It shows a mechanism de- signed to change the shuttles when any one weft thread breaks. or fails, the substitution occurring by an instantaneous move- ment, without any act of the attendant, and without stopping the loom, tlie mechanism being brought into action by a weft stopper annexed to the shuttle. The specification also refers to changing shuttle boxes to bring different colored weft into action. It also contains a jacquard mechanism. Both Reid and Johnson were prolific inventors, Johnson having taken out a patent as early as [803, lor a dressing machine, and Reid as early as 1827. for a lay motion. Johnson and Reid together took out several other patents for less interesting improvements. The discovery of the Reid and Johnson patent of 1834 displaces a former claimant ; namely, Charles Parker, who took out an English patent in 1840 for a very similar combination. The next invention in this line is of the year 18^2. Meanwhile, however, Mr. Clinton G. Gilroy issued his noted work on weav- ing in 1844, in which in a satirical and humorous vein, he refers to the loom of Arphaxad, explained to Deioces, the first king of the Medes. In the description of this loom it states : "In order to avoid stopping the motion of the loom when one or more of the weft threads break, or become exhausted, a few span- shuttles are to be lodged in suitable receptacles, which are so arrange,/ that the mere breaking of a weft thread will cause a change of shut- tle instantaneously (by the substitution of a spare one in its stead )." The detail of the operation is described at some length : also the mechanism by which the loom will stop, supposing the total number of shuttles to he exhausted. He also describes a shuttle-changer for application to different colors of weft yarn to produce patterns in the cloth. The operation of the change ot" filling is similar to that in the Reid & Johnson and Parker patents, the details seeming to show that the author was well acquainted with the Reid & Johnson patent, and possibly the Parker patent also. Gilroy's reference is merely an indirect satire on our patent system, though many of his readers have since taken this part of his work seriously. Gilroy himself was an inventor of considerable prominence in the weaving line, and must have considered the idea of changing filling too chi- merical to be practically developed. 20 Fig. 20 — Roberts Loom. >ide Elevation. In our earliest public reference to the Northrop loom, name- ly, that quoted in the paper of our Mr. George ( Hi- Draper. read before the New England Cotton Manufacturer-" Associa- tion, at their meeting in Atlanta. October. 1895, it was stated that loom- rested while improvement- changed the form of other cotton machinery, -'plain weaving remaining in its element- ary stage " Abo. "No radical change in any vital feature can be shown as the result of the last fifty years." These remarks awakened some comment and criticism, calling forth a reference in our circular, The Advance of tJtc Northrop Loom, to the loom manufactured by Richard Roberts in [830. We now .show a print of this loom, which was sixty-five years old at the time of the Atlanta meeting, and call attention to the fact that its genera] design and equipment is very similar to that of common looms at the time of the introduction of the Northrop improvements. Practically all of the important elements of plain weaving are shown in precisely the same relative positions which they now occupy : in fact, the weft fork is the only notable omission. Other authorities have since added testimony of similar sort : — "It may safely he asserted that at the present time no subject is receiving more careful consideration than that of weaving. In its essentials the power loom has changed little since the date of its inven- tion. It has heen made heavier, the details of the let-off" and the take- up and the numerous other parts have heen changed in their degree of efficiency, hut little in their method of operation. Vet from the hegin- Ding of the century it has heen clearly foreseen that a most radical change in weaving would take place upon the invention of a simple and efficient weft supplying mechanism." — [Henry I. Harriman nt tin Boston meeting <>t tin X. /■;'. Cot. Man. Assn.. April 26. 1900. The incompleteness of the earlier automatic looms is also verified : — "In the case of weft supplying looms the difficulty of transferring such a large body as a shuttle, in the very short period of time given between picks, prevented their genera] use. The process was destruc- tive both to the loom and the shuttle, and it is safe to say that none of these numerous inventions was ever put to practical use." — [//. /. Har- riman at Boston meeting of tin- X. E. Cot. Man. Assn.. April 26, 1900. "But following 187<> there was a very general absence of work on automatic looms until there appeared that remarkable series of inven- tions perfected by the Diaper Company." — [_H 1. Harriman at Boston meeting of th( X. E. Cot. Man. Assn.. April 26, 1900. HISTORY OF THE NORTHROP LOOM. In order to avoid the usually inevitable misstatements made years afterward concerning the early conception and introduc- tion of important inventions, we will briefly record the perti- nent facts concerning- the early history of the Northrop loom. The predecessors of our present Company started as far back as [816, to perfect the power loom. Ira Draper inventing the revolving temple at this period. At the formation of the partnership of George Draper & Son in 186S. the business con- trolled by this firm and other Hopedale companies chiefly rela- ted to loom improvements, including let-off motions, parallel motions, thin-place preventers, loose frogs, etc. The ring and spindle inventions, however, coming in soon after, assumed such prominence that the loom department became a secondary feature. The members of the firm, however, often speculated on the possible advantages of automatic weaving, considering this as a possible Held for future development. On July 26, iSSS. Mr. William F. Draper, Jr.. heard of a loom invention in Providence, and -aw the inventors and their device, which was an automatic shuttle-changer. lie reported at home that the general idea was interesting, hut the device not practical, in hi- opinion. Our firm then had an exhaustive in- vestigation of the patent situation made through competent counsel. The report seemed to show that there was little novelty in this special application of the idea, hut the firm had become sufficiently interested to risk a further trial of the general principle, and on December 10th voted a sum of Si 0.000 for experiment-, and started an inventor. Mr. Alonzo E. Rhoades, on the task of devising a practical shuttle-changing loom. That Mr. Rhoades lost no time is proved by the fact 23 that he had an operative loom read} to be started, with warp and filling, by February 28th of i8Sg, This loom, after being reconstructed with new patterns during the next few months. though not changed in principle, ran with good success. Some twelve years later, for purposes of patent litigation, the same loom was started up and run for days under the eye of a patent expert, accomplishing its purpose so well as to draw forth his unqualified approval. Leaving the Rhoades loom at this stage, it is necessan to retrace our history to the year 1857. when Mr. James II. North- rop was horn in Keighley, England, on May 8th of that year. After becoming an expert mechanic and factory foreman in his own country. Mr. Northrop came to this side in May, [881, soon drifting to Hopedale, where he became employed as an expert on metal patterns. His invention of the Northrop Spool- er Guide brought him to the notice of his employers, and he was selected by them to work out the idea of an automatic knot- tier for spoolers. Although showing great ingenuity, the devices did not appear commercially practical, and the inventor became sufficiently discouraged to abandon the shop and devote his time to farming. Not finding this occupation congenial, he applied for employment some years later, in the fall of [888, hut the only opening then present was a job as mechanic at $2 per day. In February, Northrop, who had noted the progress of the Rhoades idea, spoke to Mr. George Otis Draper, who had just entered the firm of George Draper & Sons, stating that it given a chance he could put a shuttle-changer on a loom in one week's time, that could he made in quantities for a cost ,,f $1 each. On March 5th. Mr. Draper drove to his farm and saw a rough wooden model of his idea, which was set up in his hen- house. At Mr. Draper's recommendation, the firm ordered an- other loom Eor experiments, and after its arrival Mr. Northrop was started on April 8th to work out his scheme. By Mav 20th 24 he had concluded that his first idea was not practical, and having meanwhile thought out a new plan, he asked for an extension of time until the fourth of July in which to perfect it. On July sth. the completed loom was running at speed, and as it seemed to involve more advantages than the Rhoades pattern, the weaver was taken off of the Rhoades loom and transferred to the North- rop. On Octoher 24th a loom with new construction, from revised patterns, was running at the Seaconnet Mill in Fall River, and more looms of the same kind were started up there- at intervals. Mr. Northrop had. however, meanwhile thought out his idea of changing filling in the shuttle, some of the parts of such a mechanism taking shape as early as October. The development at our works continued so favorably that by April of 1S90 a lot of filling-changing looms were started in the same Seaconnet Mill, the shuttle-changing looms having been changed back to common looms, in view of the additional advantages of the filling-changing pattern. To show the situation at this period we quote from a letter sent a prominent mill official May isj, 1890: "Replying to your favor of the 14th hist, would say that we are get- ting along as rapidly as we could hope or expect with our new shuttle patent, considering the fact that we are doing what seemed to be a very difficult thing and reaching out into a field where we have nothing to guide us. We are now running 12 looms in a mill constantly. They are pro- ducing from 5 per cent, to 10 per cent, more per loom than other looms in the same mill and are all making first-class cloth. We have not yet fully tested them to see how great a reduction we can make in the num- ber of weavers. This we are proposing to do at the earliest moment. We do not feel at liberty to change one or more looms for you at the present time and in explanation will map out to you our proposed course and we think you will agree with us that our policy is a wise one. What we intend to do is to perfect by practically running as long as seems necessary these 12 looms before making or trying any more. When we have perfected these 12 looms we propose to put in 100 or 200 looms and when these 100 or 200 looms are running to our entire satis- faction we shall hope to apply the invention to the entire mill. When the entire mill is running to our satisfaction we shall then be very anxious to try our inventions at other places. Our reasons for adopting this course are, first, we want to devote nil our time and energy and inventive capacity to perfect the desigJi in one place so as to be sure we shall make a success of it there. We be- llf^rf i hat in this vv;ly weshal1 be able to put the invention on 3000 or 4000 looms or 10.000 looms much more quickly and satisfactorily than in any other way." Attempts have been made by interested parties to show that these earlier trials were experimental in character, and produc- tive of nothing practical at the time. Such, however, was nol the case. These earlier trials, both of shuttle-changer and fill- ing-changer, showed practically operative mechanisms, winch were run on many looms weaving cloth lor the regular mill product, with the regular mill help; in fact, when we trans- ferred our trial of mechanisms from Fall River to another mill centre, the looms which we left were run for months by the mill help without superintendence on our part, and without even a casual inspection by any of our men. We left the twelve looms running under the normal super- vision of the mill management in March. 1891. To show how well these early mechanisms did their work we quote from the following letter received from the overseer of the room June 27. 1891 :— _ "I am proud to inform you that there lias not been a mishap of any kind tins week. The looms are weaving faster than the spinning frame can spin. Mr. seems surprised to see the weavers standing at the end or the frame waiting for the doners and their Looms stopped. Notwithstanding having to wait so many times for filling, the produc- tion for the week ending 27th is seventy-eight I 78 I cuts." ' We found it would be necessary to build complete new looms in order to derive the best results from the new mechan- isms. This required an entire equipment of the necessan tools and a considerable enlargement in plant, as we had never been loom builders. We also found that it was advisable to develop a practical warp-stop-motion for use with the filling-changer, and this of itself" delayed the introduction of the loom for sev- eral years. We ran into annoying mechanical difficulties, it requiring a long time to solve the apparently simple problem of tempering the shuttle springs so that they would not break. Even with the loom complete in every detail, vve were not reach- to take large orders until we had equipped a weave room of our own and run it continuously for mam months. To go into further detail and cover the entire ground would require more space than can now be afforded. The further eon- tents ot this hook may aid in giving a proper conception of the further development ; and yet the finished products shown con- vey no intimation of the countless experiments and trials of devices which have not entered into the accepted combination. Many of these are shown in our voluminous patents; others are still unhonored. They all form a part of the unwritten story, however, and often might furnish interesting chapters. Our manner of developing improvements is outlined in the paper of Gen. Draper delivered before the Xew England Cotton Manufacturers' Association on April 22, 1903. "Our routine lias been, firstly, to run a Dumber of loom- experi- mentally in a room in our shop, and by means of special observers, in addition to the weavers, to note results in detail. These results are collated in daily reports, which are preserved for study and reference. Notes are made of everything outside of perfect weaving - , the breakage, wear or slipping of parts, the failure of mechanism to act every lime as intended, imperfections in the cloth, like chick or thin places, the num- ber of warp and tilling threads broken and why they break, if it can be known. After studying these reports in connection with personal ob- servation of the running looms, changes are made, with a view to im- provement it possible. Pieces that break are Strengthened, or strains are removed; parts that slip are more securely fastened ; and wear is ob\iated where it seems possible. Xew devices are suggested to obviate cloth imperfections, or break- age of warp or filling, of bobbins or shuttles. The new parts are made and tested in comparison with the old ones, and nine times out of ten they don't work as well. Perhaps they don't overcome the difficulty; perhaps in overcoming it they introduce new ones. After one failure comes another attempt, and as a rule another failure, but something is learned from each trial and the general course is towards improvement. The worst troubles to find and cure are those that are intermittent and infrequent. A device will work as intended a hundred or a thou- sand times. Then it tails once from Bome unknown cause: then it goes on all right as before. <>ne seldom or never sees the failure except in result, and if it happens before one's very eyes the motions of the loom are too rapid to make eyesight of much advantage, one can only rea- son in these cases and. as in some other matters, unassisted reason with- -7 "in sufficient data comes pretty near being a guess. However, guess we lim ~' '"• let the defect continue, and in some cases we have guessed right. In others we are still guessing. After we reach what seems a real improvement on one Loom, we try it on a dozen, i -e or Less, and keep records for a month or two. Here again disappointment often comes in and we return to fresh Btudy and experiment. If. however, the advance proves real we oexl arrange a mill test ; that is, we till an order, or a part of an order, for looms with the new device, and submit it to the tender mercies of those win have to run it practically and without any special interest beyond ••day-pay and Saturday night." This kills many an infant invention that would be of value if prop- erly cared for. No oew device in minor detail can succeed in the mill if it causes extra trouble, even if it does better work : and if any new adjustment- are introduced, they are almost sure to introduce wrong set- ting. Lack of adjustment induces filing and chipping to attain positions that our experiments have shown to be wrong, but the fixers have not been through the experimenting and sometimes want to make improve- ments themselves. Cams that have been carefully worked out have been filed or ground so that they would not work as intended and the device has been condemned, and in more than one instance operating parts have been cut off with a cold chisel and the new device pronounced valueless. l After this experience we re-design, simplify and try to make the new arrangement easier to run than the old. and if we succeed and accomplish the original design, we have made a step forward. It is fair to say that from these mill tests we often gel ideas ol ^'ieat practical value from intelligent operators, who see necessities that had not occurred to us. more than enough, perhaps, to offset the stupid condemnation of others who do not appreciate fine points and never will until they have become a part of their regular drill, and only then because if they can't make a machine run. there are plenty of Others u no know how to do it.'" Perhaps nothing in the line of histon is more significant than our various statements published in the way of advertise- ments in trade papers. The whole of anticipation, progress and realization is thus set down as it was, or assumed to be, at the tune. Those that follow arc actual quotations From publica- tions of the years mentioned. 28 iS 95 . "We believe that certain improvements we are soon to in- troduce will divide the cost of weaving by two on all plain goods. We have a complete weave room of eightv looms running on print cloth, which is open to the inspection of interested manufacturers." ''It is a grave question whether we should invite more (loom) orders under the circumstances. A success may prove embarrassing when it comes so suddenly." "Textile workers should be interested in all inventions that make their labor easier, cleaner or healthier. What is more unclean or unhealthv than the now necessary process of sucking rilling through a shuttle eye ? We are introducing a loom which automatically threads the shuttle without labor on the part of the weaver. This loom also prevents damage to the cloth, caused by broken warp threads." "Many persons are disappointed in the Northrop Loom because it does not produce finished goods at one end from a bale of cotton led into a hopper at the other side." "We believe a purchase of common looms a grave error at the present day." 1S96. ••A mill that orders common looms at the present time de- liberately handicaps its future prospect-." "We now recommend this (Northrop) loom and stake our reputation on its success." "The majority believe in progress. They favor inventions that relieve human labor by transferring operations from fingers to levers and cams. The Northrop Loom is of this class." "We do not have to reply on assertion. Thousands of (Northrop) looms are in actual use testifying to their own merit." "We have had additional orders already from six of the first ten mills supplied." "Consign your common looms to the scrap heap where they belong, and equip with machines that will earn a profit." 1897. "The Northrop Loom is now an Unquestioned Success on all plain cotton fabrics. . . . We have never had a more positive conviction. This Loom must be adopted." "When mills like the Pacific and Tremont & Suffolk throw out common looms for New Northrop Looms, the question of success is solved. Before the year is over the Amoskeag Mfg. Co. will have nearly 10,000 looms changed to take our motions." ••Weavers on all common looms choke their lungs with cot- ton fibre. When the filling is colored the effect is more or less poisonous, and in either case the health is undermined." "It is commercial suicide to buy a common loom in the face of facts easily known and proved." "Why not return to hand looms and get a cheap equipment, also giving more- laborers employment?" 1S98. 'What would you think of a loom that requires but half the labor, weaves more perfect cloth and will run over time without need tor attention? \\ ould von buy it at a price that makes it the cheapest ma- chine ever put in your mill, or would you wait, and doubt, and doubt and wait, until the competition ol" the enterprising forced you into line at the rear of the procession?" "Adverse criticism has often killed a good idea in it- infancy while it- strength was not equal to the struggle. We escaped the fate which many prophesied." ••The only hope for our cotton mills in these critical times lies in the prompt adoption of improved machinery. It may he urged that if all mills put in new machinery they will -imply be back at the old competitive level — very true — but they will not all do it. Therein lies the chance for profit for those who have the necessary courage, capital, or happy combination of both." •"The doubters and the skeptics are not yet silenced — they never will be. Some of them still think it a great mistake for mills to use high -peed spindles, filling frames and revoking flat cards. We have no time to waste on their conviction, as their species must yield to the natural law — the survival of the fittest." 1899. "The mills that refuse their opportunities will find their future utility serving a- picturesque ruins in the landscape." ■•If old mills stand in timid dread on the brink ol' indeci- sion the new mills will crowd them over the edge." "You can feel assured that merit is recognized when the copyist appears — but you don't want a cop} ." "Let us then renew the assurances of our distinguished con- 3' sideration, while we devote our energies to rilling the orders with which we have been Eavored." 1900. "The greater part of the cloth woven in this country is made on plain looms. We have devoted about 10 years to the perfection of the plain loom and have now made it automatic and self-protecting against errors." "We intend to keep up with the demand for our machiner) if we have to roof in the whole town." "A new common loom in a Southern cotton mill is now a curiosity." "We are battling with nature, filling ponds, diverting river channels, raising valleys, etc.. to make room on which to con- tinue extensions." "We still solicit orders in the confidence that bricks and lumber may he obtained in sufficient quantity to house our in- crease of plant." ••Win' ship cotton to Europe when mills at home can man- ufacture it moie profitably now that improved machinery gives them another advantage?" '•The great development of the Southern cotton mill system started with the Northrop loom and the continued association ot the two tonus an interesting object lesson." "We melt 100 tons of iron per day to make the castings for our Northrop looms, etc. But that is not enough. Enlarge- ments still in progress." ••We have now sold over 60,000 Northrop looms. We are shipping 1 soo a month and enlarging our works to increase that output. We are employing 2^00 men and shall great!} increase 3 2 this force when new shops are ready. And what does this all mean? Simply that the success of the Northrop loom is astound- ing, even those who have held their faith." "The steady progress of the Northrop loom is a certain evi- dence of its merit. Adverse criticism has often killed a good idea in its infancy while its strength was not equal to the strug- gle. We escaped the fate that many prophesied. Our loom has passed the trial stage." "Let all who favor progress unite in placing American cot- ton mills where they can compete with foreign countries without reducing their labor scale to the standard set in England, Ger- many, Russia, India, China, Japan and other outside manufac- turing sections." "We build the famous Northrop Loom. It is also manu- factured by our licensees in Canada, Germany, France and Switzerland. Four of these looms are running at the Paris Exposition, attracting wide attention." "The successful development of our loom gives a mill a chance of making a great saving in its expenses without increas- ing the labor or responsibility of the management, and by reduc- tion of the number of employes it actually lessens the invest- ment necessary for tenements and the labor used in paying off and supervising. The possible profit from a Northrop loom mill will pay good dividends when a competing mill with com- mon looms is not able to show more than an even balance. Mills have been prompt to take advantage of improved machinery in the past, as they universally use high-speed spindles and are thor- oughly committed to the revolving top flat card. Neither of these changes, however, can show more than a fraction of the profit possible with our loom, for the saving in weaving is more than the entire cost of carding with the picker-room thrown in, and more than the entire cost of spinning." "New mills are flooding us with orders, and old mills must 33 realize that equality in competition demands equality of equip- ment." "We used to claim that weavers could attend Northrop Looms in the proportion of two to one common. The users are finding this prediction far too moderate as they often run three to four times their former limit. In several mills weavers are paid less than one-half the former price for weaving cloth pei- cut, and yet make higher wages than when running common looms. A mill that cannot appreciate that statement simply cannot appreciate the tale told by concrete figures. Those who attempt to sell cloth handicapped by an extra cost of from one cent per pound upward, can cling to their obsolete common looms while their more enterprising neighbors glean the profits." 'AVe begin to feel quite independent in our loom trade, as the results of experience have proved that our position is abso- lutely unassailable. A few facts speak for themselves: (; ()() ,1 weavers running 24 to 32 print looms and 20 3-harness looms." "In one large print mill the average number of looms per weaver is 18." "We are employing more hands than ever worked before in an American Cotton Machine Shop and are enlarging our plant in every direction." "Every new idea meets the same opposition, goes through the same routine. In the first few years this machine had to bear the brunt of criticism, antagonism, doubt, fear, and mis- representation. Now it suddenly sweeps away opposition, flood- ing us with orders, and necessitating the doubling of our plant. We intend to keep abreast of the demand if pig iron and steel can be obtained in sufficient quantity." "It is an interesting problem to note how much longer the old mills can continue competition, when handicapped by the obsolete common loom." 34 1 90 1 . "With a record of 7^.000 looms sold, it is no longer neces- sary for US to predict what these looms will do. We point to what they have done." "Although our order list lengthens and strengthens, we do not adopt the simple and inexpensive plan of building without change, but continually add improvements whenever possible." "We shipped more than 16.000 complete Northrop looms during the last year. What better testimonial of value could be presented? With our new plant and enlarged facilities we shall easily beat that record in 1901. This simply means that those running common looms must expect a continuously harsher competition." "Having adopted a business founded on improvements in cotton machinery, the habit of striving after perfection leads us, at times, to give the public more than they have required. Al- though the Northrop loom has sold faster than we could supply the trade, we have recently made main expensive changes, in spite of the fact that our customers, if ignorant of their existence, would probably have never realized the need of them. All loom improvements tend toward increased cost of construction. We have taken the common loom and not only applied import- ant attachments, but have also raised its mechanical grade." "Every loom that we sell furnishes an additional argument for replacement of common looms, as each Northrop Loom in- creases the competition that its rivals must endure. Those having common looms must admit that, sooner or later, the Northrop loom, or some similar type, will replace them. Then why delay? Every year of postponement could have helped to pay the cost. Those who are waiting for the similar type to be developed can hardly find a large degree of encouragement from the present situation. They used to wait, in the same way. for spindles of possible competing capacity in 35 earlier years. They waited five, ten, twent.3 years, — and then finally Eel! Into line, after losing a large share of their compara- tive value. Some of them lost time and money in experiments with inferior styles, and history will undoubtedly repeal itself. Some insist on patronizing cheap doctors, cheap lawyers, and cheap eggs. Perhaps they are satisfied with the results. Our loom is not cheap in price, hut is certainly cheap at the price." ••Idle success ot the Northrop Loom has forced a series ot wide spreading events. It has delivered the trade in looms, for plain Fabrics, of the United States, over to a company which had never sold one loom prior to [895. It has stimulated the building of new mills and the increase of the American textile industry to an extent never before known. It has forced us to more than double our plant, and more than treble our number of operatives. Idle profits have been shared with the manufacturer, who has cheapened production; and by the laborer, who has received better wages. While common loom mills are shut down, Northrop loom mills continue running." "We shipped more than 2^.000 complete Northrop looms during the eighteen months of January, 1900, to July, 1901. What better testimonial of value could he presented? Southern mills are taking their share, hut there are still thousands of old looms that ought to he replaced." "We shipped nearly 6,000 looms in the first three months of the \ear 1901 . Facts like these carry conviction to those ot average comprehension. We shipped over 9.000 Northrop looms from our works in the six months ending July 1. 1901. Further com- ment is unnecssary." "We enter on the seventh \car of our loom business with 36 an enormous order list, a doubled plant, and a reputation estab- lished by the experience of our customers. Every claim ha- been justified, every assertion proved. The Northrop Loom does halve the labor cost of weaving, make better goods, and does earn dividends for it- pur- chasers. Having absolutely removed the common loom from compe- tition, so far a- new -ale- are concerned, we may next have to spare some -light consideration for the mushroom element of automatic substitutes designed to -hare the fruit- of our victory. Let none of us yet unduly excited, however, until their trial has proved them worthy of attention." "We -tailed to apply attachments to loom- in order to make them more automatic. We soon found it necessary to first improve the loom itself. We believe that we are turning out a weaving machine fit to class with other developed mill machin- ery, and not a cheap ma-- of ill fitting part-, half wood, half metal, nursed into efficiency with bits of leather and string. Our castings are machine moulded to ensure uniformity. They are drilled in jigs and assembled to gauges. We use iron and steel wherever possible. We know we put more expense into this loom than any other builder of similar machine-. We are not content with having already done a larger loom business per year than any competitor. We see no reason why we should not -ell all the loom- needed for plain weaving." "Our total -ale- to date, including old loom- changed over, amount to over 74.000. We have built up a modern plant of large capacity in order to meet the demands of our customers, and now have 22 acres of floor -pace in connected buildings, the greater part of which represents recent construction. \\ e are now ready for increased business and await it with a confidence based on the evolution of the pa-t. It may be 37 noticed that we refer more often to the amount ot our sales than to the details of our products. The latter course would siinpb illustrate <>!/>■ opinion, while sale^ illustrate the opinion of our customers — and that counts." "We know no halt-truths in mechanics. A machine is either efficient or incapable — superior or infe- rior. The Northrop Loom has now - been running in large quanti- ties for more than six years. Its success is proved by the fre- quency of orders from those having the knowledge that comes with use. Some of the earlier customers have lately wished to actually duplicate those first machines part for part. But we build a better loom now. We have an experience gained by continued construction and experiment. We have vastly increased our range and our variety of models. We cannot only show a purchaser important novelties, but can refer to successful operation in any of the ordinary lines of application." 1902. "The largest single order we have yet taken has just been placed with us for Northrop Looms by the Grosvenor Dale Co., of North Grosvenor Dale (and Grosvenor Dale), Conn. These looms were chosen after lengthy and continued trial of former lots. These were used in a wide variety of cloth, including various standard weaves for which the Grosvenor Dale Company has long been famous. Those who have been cautiously awaiting the outcome of others' experiments may now perceive the verification of our earlier contentions." ••The Spindle and the Loom. Our first ten years of spindle sales, about 2. 000. 000. 38 Our thirty years of spindle sales, about 20.000.000. Every prominent mill in the country uses them in their Spinning Frames or Twisters. And vet in the first ten years the introduction was compara- tively slow. Our first seven years of loom sale- figure over 7^.000 (in- cluding looms changed over), and there are only about 37:5.000 looms in this country to which our improvements are at present adaptable. Every mill that waited to change spindles made a mi-take. They admit it by their present policy. A less proportion are making the mistake of indecision in the loom line, but the conservative are still ruining their chance in the same old way. Every year of delay means just so much lost profit. The above figures of fact prove more than pages of argument. Think them over." "On June 1st our unfilled orders for complete Northrop Looms figured exactlv 15.701 — and the boom has hardly started." "Our unfilled orders for complete Northrop Looms figured exactly 21,586 July 1st. 1902. The boom is beginning to boom." "Delegations of foreign business men. operatives and labor leaders have been visiting this country to investigate the claimed advantages of our Northrop Loom. We started selling them eight years ago and have averaged sales of over 10.000 per year. Outsiders are becoming alarmed and yet there are American mills still blindly Inning common looms. Not that we have any reason to complain. It takes a doubled plant to keep pace with our orders — but it ought to take a trebled plant." "In spite of loom shipments during August of 1799 looms, 39 our unfilled orders still amounted to over 20.000 September 1st. 1902." "Out of 64,540 looms now running or ordered by the single state of South Carolina, 27,980 are Northrop Looms." "20,000 looms to build. 20,000 Northrop Looms. Equiv- alent in cost to 60,000 common looms. 10 months' work at 2,000 looms per month and new orders coming in all the time. Works must be increased again. 300,000 looms yet to be replaced in the United States alone, and new mills being organized. Such is the situation confronting the Draper Company of Hopedale, Mass." 1903. "We shipped 15,746 complete Northrop Looms in 1902. and applied besides, 1.028 rilling changers and 1,234 warp-stop devices to looms in mills. We commenced the new year by shipping 2.^00 complete looms the first month. Let the good work go on." ••Our present output of Northrop looms, over 2.000 per month. The majority of new orders are placed 1>\ Southern Cotton Mills." ••\\ e have today sold over 80.000 complete Northrop looms. We have applied attachments, in addition, to over 1^.000 looms. We figure that there are still 350,000 looms that must he replaced. They will vanish as surely as the common spindle and the old style card. We are enlarging our plant t<> prepare tor their elimination. In a certain well known mill six weavers and four boys to till hoppers run 216 Northrop looms. In an- 4 o other mill no weaver runs less than 24 Northrop looms. Facts like these breed conclusions." "We have a new Northrop Loom that should be of interest to weavers of print cloth and similar goods. It has the latest large pattern hopper, our steel-harness warp stop-motion with simplified knock-off, a double fork to prevent thick and thin places, the simplest take-up ever devised, our improved Draper-Roper let-off. and a new device called the Anti-bang, which prevents jar and breakage when a shuttle is trapped. We call it the J model. Large orders already being filled." 4' THE PRESENT STANDING OF OUR LOOM— APRIL, 1904. A record of over 100.000 looms actually introduced within a period of nine years, sold at prices equivalent to three times the cost of the common looms with which they compete, is cer- tainly sufficient evidence that the Northrop loom has come to stay. The amount thus paid us for Northrop looms would actu- ally replace three-quarters of the common cotton looms now running in the whole United States. As our last year's sales were larger than those of any previous year, it is evident that the introduction is not based on any quick enthusiasm, or false data. We started with the assumption that the Northrop loom would enable the weaver to produce a doubled product; in fact, before even making this modest assertion, we proved its truth to our own satisfaction by running a weave room of eighty looms in our own works, for many months, open to the inspection of hundreds of practical mill men. The first looms that we put out were therefore seasoned, as it were, by experience: in tact, the first models ran so well that we have been asked in recent years to duplicate them. It is no slight task to introduce an improved machine which aims at replacing the entire equipment of the most important section of one of the greatest industries in the world. It cannot be done in one year, or one decade. Nothing within our mem- ory has so completely misted competition as the high speed spin- dle; and vet comparison of sales will prove that our loom has met with readier appreciation in the earlier years. There are still several hundred thousand common looms which should be replaced, and which will be replaced. The delay is not due to 4- hesitation based on disbelief, but rather a hesitation based on financial conditions. With new mills, where capital is raised by subscription, equipment with Northrop looms is becoming a mat- ter of course: hut an old mill faces a serious proposition when considering the replacement of an entire division of its plant, where the surplus is not sufficient to meet the cost, and where stockholders are not inclined to pay assessments, or take new stock. The mills that have a comparatively new equipment of common looms are naturally indisposed to reduce their valuation by considering them practically worthless for active use. We are. however, selling tons of looms for junk, that are equal, if not better, than similar looms still bought by a few obstinate adherents to ohsolete methods. There is also a class of overshrewd managers who wait in hope that competition may reduce our prices, or that patents will expire in time to force a reduction to meet their demands. Nine years of constant introduction rinds the anxious ones still waiting the possible competitor; and the con- stant improvement, with continual issue of important patents, assures us that our hold on this line will continue beyond the time to which their hopes might limit us. Meanwhile these waiting purchasers are losing the possible profits of use. The fact that they may lie making favorable showings by reason of ••luck" in purchase of cotton, especial advantage in situation, labor, or power, cannot disguise the fact that with the Northrop loom their profits would be still higher. "\\ hen we refer to the Northrop loom improvements, we are speaking primarily, of the filling-changer, the warp stop-motion, and their co-operating parts. Before our application of these devices, there had never been a successful use of filling-changing devices of any nature, and warp stop- motions were only used in a very limited held, a few in- stances being known of their application to special classes of double warp-weaving. There is hardly any vital change in any 43 line of mechanics, which so suddenly brought successful auto- matic mechanism into extensive use, without the preliminary record of long use of partially successful devices of similar nature. This fact is particularly curious, in view of the fact that a warp stop-motion was one of the inventions disclosed in the original power loom specification of Cartwright, as shown in his patent of 1784. Main- inventors had struggled for years with the prob- lem of automatic change of shuttles. The inventor of the Nor- throp filling-changing devices, however, borrowed practically nothing from the former art in this line, and when it was found necessary to incorporate a warp stop-motion with the filling- changer, there was nothing formerly developed that could he adopted, and inventors practically started in this Held also with- out the aid of prior thought; Neither the filling-changer nor the warp stop-motion neces- sarily increases production in the loom itself. The filling-changer does save time formerly occupied in changing shuttles by hand, with the loom stopped, hut the warp-stop-motion actually decreases production by stopping the loom oftener than it would he stopped in the common practice of plain weaving. The com- bination of the two devices, however, allows the operative to multiply efficiency : for tin- filling-changer replaces labor, and the warp stop-motion relieves the annoyance of constant oversight. To appreciate the great saving introduced by the filling-changer, it may he well to note the operations gone through by a weaver on a plain loom, when the filling is exhausted. They follow in the sequence now recorded, the weaver performing the following functions : 1. Releases the shipper brake. 2. Pushes the lay hack. 3. Withdraws the shuttle. 4. Puts the reserve shuttle in the shuttle box on the la\ . ^. Pulls the shipper handle to starl the loom. 44 6. Rubs the cloth below the breast beam to prevent a thin place, if light goods are being woven. 7. Picks up the discarded shuttle again. S. Pulls the shuttle spindle out on an angle. 9. Removes the empty bobbin or cop tube. 10. Puts in a new bobbin or cop. 11. Pulls off a sufficient length of filling. 12. Snaps the shuttle spindle back into place. 13. Holds the filling over the shuttle eve entrance. 14. Sucks the filling through the eve. 15. Places the shuttle in its holder, where it remains until needed. Now, this series of performances must be gone through with every time the filling is exhausted. On one loom, the filling may run from one minute to twenty minutes, according to the size of the yarn and the amount of yarn in the shuttle. The average time is perhaps six minutes, especially if we count the number of times that the weaver must come to the loom to start it up when the filling breaks. With a loom having an average of six min- utes between such stops, the weaver must come to the loom once every ten minutes. If running eight looms, he would have such a duty nearly once a minute. With the Northrop loom, on the contrary, the weaver can fill a hopper containing 2^ bobbins. which, with the same average of running time, would last two hours and a half, without requiring attendance. But a co-opera- ting feature of great advantage with the Northrop loom is the fact that the weaver can fill the hoppers when convenient, rather than be forced to come to the looms with irritating regularity. Referring to the associate attachment, the Warp Stop-Mo- tion, it is. of course, well known that the warp threads will break in weaving. On a common loom, the broken thread will not be raised by its heddle, and thereby leaves an open space in the cloth, more or less visible to the eye, according to the character 45 of the goods woven. Very often the broken end gets tangled around adjacent threads between the harness and the reed, hold- ing several of them cither above or below the tip of the shuttle, which therefore causes a defect known technically as a "float" or "overshot." If the weaver does not notice' the tank promptly, the extra strain will break main of the warp threads, and in any event, a pickout is necessary. In some mills, a weaver is forced to stop all looms under his charge while attending to a pickout. It is not necessary to explain the trouble caused by these defects to any weaving expert. The temples must he pulled back, all the filling threads that have been laid since the tangle commenced removed by a tedious combing operation, the warp beam must be turned back, the tension of the cloth properly adjusted, and the loom again set in motion. When we first applied filling-changing devices, we found that the weaver, although greatly relieved of manual labor, was even more uneasy, on account of possible overshots, having more looms to look after. We saw that it was absolutely necessary to furnish a protection in the way of an accurate warp stop-motion, so that there should be no mental anxiety whatever, and no neces- sity tor alert observation. It took our inventors several years to produce a practical mechanism of this nature: in fact, the intro- duction of the Filling Changer itself was delayed for quite a period while waiting for the associate mechanism. With the protection oi the Warp Stop-Motion, a weaver is only limited in the number of looms attended, by the amount of warp bleaks which must be repaired, and the number of bobbins which can be put into the hoppers within the time to be given. Under present s\ sterns^ Northrop loom weavers are usually relieved of oiling and cleaning their looms, so that apart from the warp and filling duties, they have practically nothing to attend to. save the re- moval of the cloth. 4 6 PERSPECTIVE VIEW OF EARGE HOPPER, ORIGI- NAL DESIGN. Taking the various attachments in order for detailed consid- eration, we shall consider the Hopper first, as the more import- ant element of the whole combination. The cut shows the "Large Hopper," or "24-Bobbin Hopper." It is operated in rotation by the reverse motion of the transferrer. Our latest forms have a new and improved bobbin support. 47 r provided with a leather washer cushion to prevent breakage, and we are also using a new form of bobbin tip holder, which will take either bobbins or cops, as desired. We start- ed with a hopper that held a supply of 14 extra bobbins, but the change to the 24-bobbin hopper has proved a distinct ad- vantage, removing the Northrop loom still further from competi- tion with the possible perfected Shuttle-Changing loom, which would probably be limited to a reserve supply of six or eighl shuttles. We proved by an absolute test on our old hopper, that a reduction in the number of bobbins held in reserve, placed an absolute restriction on the weaver's capacity; and the converse of the proposition is a natural sequence. A vital principle of the Northrop invention is contained in the Shuttle, which is adapted not only to hold a bobbin or cop skewer, but to hold it so that it may be automatically removed by the entrance of a new bobbin or cop skewer. The spring jaws of the Northrop shuttle co-operate with the ring- or ribs, on the bobbin or cop skewer. SO as to hold either one normally 4 s Our h i- s t Large Hopper. Holds twentv- four extra bob- bins. Is rotated by reverse action of transferrer. -tzr&nn^ Present pattern Large Hopper. New end hold- ers adapted for either bobbins or cop skewers. Also new bobbin sup- port, and thread discs with wider surfaces for thread to hear against. 49 in proper horizontal position, and yet liberate them quickly when opened by the entrance of the new filling-holder, pressed into the shuttle from the hopper when the transferrer is in motion. The Transferrer is a simple pivoted lexer with a hammer head, normally in position over the lowest bobbin in the hopper. A pivoted dog attached to a crank arm on the transferrer is normally out of reach of a moving part on the lay called a "Bunter." When the supply of filling in the running shuttle is either broken, or exhausted, the ordinary weft fork de- tects the fault and by simple co-operation with a moving shaft raises the dog aforesaid to meet the impact of the hunter, therein transfering the forward movement of the lay through the trans- ferrer pivot, to press the transferrer head down onto the reserve bobbin in the hopper, and push it into the shuttle. The bobbin formerly in the shuttle falls through the exit opening of the shut- tle, down onto a guiding chute into a large box, or receptacle, attached to the loom side. Reference to the cuts, which show various views of the shut- tles, bobbins and cop skewers, will make the operation clear. It is not only necessary that the new bobbin should he placed properly in the shuttle, hut it is vitally necessary that the thread on the new bobbin should enter the shuttle eye, SO that it ma\ he properly drawn off in weaving. The threads of the bobbins in tlie hopper are wound round a stud in the center of the rotating hopper itself: and when a bobbin is transferred to the shuttle and the shuttle i>> thrown by the picker-stick, the thread still held by the hopper disc automatically enters the slotted eye of the shuttle : the final position, however, not being attained until the shuttle has been thrown hack from the opposite side of the loom. It is quite evident that when the shuttle receives the bobbin it must he under the hopper in approximately correct position. Variation is allowed by reason of the several notches in the shut- tle spring, and also by reason of an incline, which guides the 5° bobbin down into the spring, even if the- shuttle be quite ;t dis- tance out of place. To protect against any abnormal position, which would cause the incoming bobbin to strike a solid part of the shuttle and cause breakage, we provide a device known as the "Shuttle Position Detector." which reaches a finger across the front of the shuttle whenever the dog on the transferrer is raised. If the shuttle is in the path of this finger, the dog will not be raised sufficiently to encounter the hunter, and therefore no transfer will take place. If this be twice repeated, the loom stops automatically by a de- vice called the misthread stop-motion, attached to the fork slide, so that the weaver knows that the shuttle is not being properly picket!. The same mechanism will also stop the loom, pro- viding the hopper is exhausted, or pro- viding the shuttle fails twice to thread, or "misthreads," as we term the opera- ation. It will be remembered that the thread on the incoming bobbin is at- tached to a stud on the hopper. It therefore extends from the stud to the cloth, and. if not attended to, would break in time from the strain, as the cloth moves towards the take-up roll. Early form of Thread- and the snapped end might fly into the Cutting Temple. cloth. We therefore provide a Thread- Cutting device, attached to the regular temple, which operates from the motion of the lay to sever any such threads close to the selvage. As it operates every time the lay beats forward, it has many chances to cut the thread. SI Later form of Thread-Cutting Temple. Made with solid heel SO that a loose heel will not bring extra strain on the cutter and cause the temple to be reciprocated through the cutter. It may be noticed that Filling-Changing mechanism includes five distinct and separate devices, namely; the Filling-Changer itself, the Shuttle, the Shuttle Position Detector, the Misthread -Stop-Motion and the Thread-Cutting Temple. There are. therefore, several distinct lines, all covered by patents, many of which extend much longer than the original patents on the origi- nal mechanism. 53 As shown, the transferrer is placing the bobbin in the shuttle, the dog being in engagement with the hunter on the lay. The empty bobbin is falling down the chute into the box. The chute as first designed was a movable part, independent of the lay itself. This cut was made from our first model loom and happens to show the hopper on the left side, we making hoppers in rights and lefts at that time. 54 CROSS-SECTION OF A NO. 1 COP-HOPPER WITH TRANSFER TAKING PLACE. The entering cop skewer has just started the pressure that expels the one in the shuttle. It has still to move some distance down the chute to reach the box. The expelled skewer is not empty in this instance, as it illustrates a case in which the rilling thread broke while weaving-. 3.1 DETAIL OF CONNECTION Between the filling fork which detects the absence of filling and a No. i cop hopper or magazine. 56 ILLUSTRATION OF HOPPER ACTION ON B MODEL LOOM. 57 A is the filling-fork which detects absence of filling, and through the usual catch and vibrator gives action to rod B con- trolling latch C. When latch is raised it will be in contact with a hunter on the lay, thus forcibly depressing the transferrer D which pushes a bobbin from the hopper into the shuttle beneath it. at the same time expelling the one carried by the shuttle, which is then guided into the large tin box held on the loom side. E is a portion of the device which determines the position of the shuttle in the box. If not properly in place the latch C will not engage the hunter, as the device of which E is a part will be prevented from further movement by contact with the shuttle tip. and as E and C work in unison, the movement of C is also checked. This special shuttle position detector did not go into extensive use on our own looms but was adopted as standard bv our Canadian licensees. 58 - A FEW OF THE SHUTTLES USED WITH OUR NOR- THROP LOOMS. Shuttle at the left is k n o w n as t h e "Keeley," although in- corporating the spring of J. II. Northrop and the incline cover of (i. (). Draper. Shuttle at right is known as No. 7 in our shop records. It has what we know as the "Stimpson" e\ e. This Hrst model had no friction pocket and the eye casting was held hv a nut on the In itt< )in. € i 59 1E1 V n No. 16 Shuttle. Stimpson improved e\ e with pocket for Manuel friction, the casting be- ing held in the shuttle body by a transverse bolt and nut. Various mod- ifications have other numbers, but this is the regular standard de- sign which has gone into most extensive use. No. i 6 Shutth Stimpson special eye and new spring and cov- er. A very satisfactory model. Note the new spring and cover which leave the wood of the shuttle body less cut out and therefore stronger. We have little trouble with loose springs in this design. 6o Xo. 263 Shuttle, or Jonas Northrop Eye stvle. Very successful on cops and coarse til- ling; in fact, the best threading eve which we have for all classes of work. This shuttle saves rilling breakage ami makes misthreading immaterial. We recommend it unreservedly until a better design is possible. 6i NORTHROP LOOM BOBBINS AND COP SKEWERS. No. i . Represents our early type of bobbin — now discarded. No. 2. First abandonment of groove at base. No. 3. Standard pattern with Claus step for ordinary yarn. No. 4. Special Queen City pattern. No. 5. Standard for coarse rilling. Note also the base on this and No. 6, made larger than formerly. We tit all our new looms to take this bobbin, as it has greater strength and is less affected by reaming. Of course, all the other patterns can have this same base. In ordering filling spindles for these bob- bins be careful and specify the large cup. No. 6. Special metal base cover extending under bobbin rings to prevent their loosening. A most important improve- ment. 62 & — ' No. 7. Feeler bobbin for use with feeler or mispick pre- venter loom. This style has three rings on base. Note cham- fer on rings at ends. We have all our rings made in this way now. as they are less liable to catch yarn in the spinning-room. Xo. S. Long traverse bobbin, special notches on barrel. No. 9. Long traverse bobbin with ordinary ribs. No. 10. Cop skewer. 63 Our bobbins and cop skewers are made in three lengths, i. 6 3-4 inches long for traverse of s 1-- inches. 2. 7 3-8 " " •• " » 6 i-S " 3. S " " " " » 6 3-4 " The exteriors shown in the cuts are used on all three lengths. We have many additional contours to suit the whims of customers, but those shown are approved by use. All bobbins and cop skewers must be ordered from us. They are patented articles. \\ e insist on this simply to protect the successful operation of our looms. We do not take profit enough to pay us for the trouble in handling this part of the business. '•A few nights ago the night watchman of the mill told of seeing- strange sights and hearing queer noises during the small hours of the morning. He is a sober man of middle age and in perfect health, so it was hard to find reason for not believing his story. He says that shortly after midnight he heard a noise in a remote corner of the mill like the running of weaving looms. He went there and found six looms running at full speed without any apparent motive power ami cloth was being' woven without any guidance." — [/row dis- patch to the New York World, Nov. 17. 1000. "The Northrop-Draper loom has had many tests and made many records. We will now chronicle one that, in romance, sur- passes the loom of this make at Tucapau mills. Wellford, S. C, winch ran nearly i\ hours without stopping a second : — Young couple engaged— against wishes father— hurried con- sultation—wedding party gathered in the dynamo-room— returned— the bride finding all her Northrop loom- running along a- merri- ly as ever."'— [Textile Excelsior. 64 STEEL HARNESS KNOCK- OFF. This cut illustrates a harness warp stop-motion as used on some of our looms. The vibrators are rocked from a cam movement on the lower shaft ami the trip acts directly on the shipper. 65 WARP STOP-MOTIONS. At the start of our loom introduction, we limited ourselves to the weaving of two-harness goods, utilizing simple warp stop- motion devices, which were perfectly efficient in this field. When we began to supply looms to weave with 3, 4 and 5 har- nesses, together with the Held covered by dobbies, it became necessary to develop new designs, so that we now have four dis- tinct styles of warp stop-motion, and mollifications in each class. Whenever possible, we recommend the use of our steel harness stop-motion. This has only been adapted to more than two har- ness work in recent years. With this arrangement, the heddles themselves serve as warp-stop detectors, heing thin, flat steel ribbons, sufficiently stiff to act in arresting the motion of a vibrator. The heddles are strung on bars, through slots much wider than the bars themselves; thus when a thread breaks the heddle may drop a distance equivalent to the extra length of the slot, and thus come within the path of a moving vibrator which, when arrested, effects the stopping of the loom by inter- mediate mechanisms. The advantages of the Steel Harness Warp Stop-Motion over all other kinds are numerous. In the first place, the heddles themselves are practically indestructible. They show no signs of wear after years of use: in fact, they are probably better for use, through the polishing given by the passing threads. The cotton harness, with which they compete, wears out, needs revarnishing and probably averages an expense for repair and replacement of perhaps a dollar a loom per year. We see no reason win the steel harness should not wear at least twenty years, saving some nineteen dollars in actual outlay, if our premises are correct. Another important advantage, especiallj noticeable with coarse yarn, is the saving of expense in drawing-in. Cotton harness 66 warp stop-motions, with additional warp-stop detectors, cost more to draw in, because the drawing-in hand lias to draw threads through the detectors as well as through the reed and harness. Our steel harness is even easier to draw in than the cotton har- ness, for the heddles may slide on the bars at will, accommodat- ing themselves to the convenience of the operative. Another ad- vantage of the steel heddle warp-stop is that it will stop more promptly, preventing warp runs after warp breakage : and itd not stop so often for slack threads. A further advantage, of great importance in mills where they change the product frequently, is that the steel harness heddles space themselves automatically . - that the same harness may be used for various weaves. The free lateral movement also allows the weaver readier access when repairing broken warp threads. We have been asked more than once why it is that the shut- tles in a Northrop loom fail to throw out of the loom like the common shuttle: in fact, our shuttles stay in the shade so uni- formly as to question the need for shuttle guards. The reason i-- easily seen on investigation. Shuttles are thrown out of looms for several possible causes, hut the most frequent one is the formation of floats, or the preliminary to a pickout. All of our looms will stop before a float can make serious trouble, and our steel harness warp-stop type will stop the loom before the warp threads can tangle sufficiently to swerve the shuttle from its proper course. In the line of steel heddle warp-stops we are absolutely with- out competition. No other loom builder has ever attempted to introduce this class of devices, to our knowledge. In the earlier use of the steel harness, it was claimed that the steel heddles broke more warp threads than the twine har- ness. This max have been true at that time: vet the advantag - were more than enough to compensate. After learning" proper methods of sizing, proper shape of cams and proper arrange- 67 r ^ 3 n x« ment of heddles in their frames, we have now broughl the steel heddle warp-stop where it is practically equal to the cotton harness in the number of warp faults. On regular print weaving, we find the stops for both breaks ami slack threads combined, is between 10 and i^ per day, per loom, with either steel heddles or cotton harness. At present, we do not sup- ply steel harness mechanisms for a greater number than five harnesses. The original cotton har- ness warp stop-motion which we introduced, used a drop wire device applied between the heddles and the lease rods, each detector serving for two or more threads. This could he used on looms having more than two harnesses, in a large number of applications. This motion was \ er\ successful and has been used on thou- sands of looms. W e are re- cently applying a stop-motion situated between the lease rods and the harness, hut which uses one detector for each thread, which looks \er\ promising for cotton harness work with anv number of harnesses. No. 2 m No. I Steel heddle. No. 2 Cotton harness drop wire for "Roper" warp stop. No. 3 Detector for single- thread stop-motion. 6S A fourth type is known as our Single-Thread Warp Stop- Motion, and has been very largely used in recent years. In this class, one detector acts for each thread, and a peculiarity is noticed in that the detectors are arranged in two banks, and placed in the position of the usual lease rods, where they ac- complish the functions of leasing- devices, a- well as warp stop- motions, doing away with the necessity for leasing rods, and simplifying the loom to that extent. Where drop wires are applied hack of the usual lease rods, a broken warp thread does not always promptly allow the drop to operate, as the lease rods sometimes make sufficient friction on the thread to hold the drop in position. While we are subject to more or less competition in applying warp stop-motions to old looms, our competitors are either lim- ited to use of electrical devices, with their inherent evils, or to the use of warp stop-motions in which the detectors are subject to a more or less severe twisting strain. Our patents cover the use of serrated vibrators which can engage the detector- without twisting and bending. Sometimes the vibrators and co-operat- ing devices on competing devices are made light and delicate. in order not to bend the drops, and therefore are less positive in action, and more liable to damage. So far as the application of warp stop-motions to other than Northrop looms i- concerned, we were interested primarily in applying warp stop-motions to looms that could not use the filling-changer, such as drop box looms. We have taken little interest in attempting to introduce warp stop-motions on common looms for plain weaving, because we consider such application a mere makeshift, in view of the greater advantages of the combined filling-changing and warp- stop, which the mills should avail themselves of. rather than attempt to try and cheapen their weaving by adding expensive devices to old machinery. Warp stop-motions of themselves, do not lessen the weaver's labor, except in the prevention of 6 9 Moats and overshots. Every thread that breaks must be pieced up, as formerly, and it is even possible thai the additional weigfht of the detectors causes more breakage. DEVICES FOR MAKING PERFECT CLOTH. The third new attachment introduced with the advent of the Northrop Filling-Changer and the Warp Stop-Motion, is a mech- anism only used on certain classes of goods, which co-operates with the filling-changer to prevent mispicks, and thus make per- fect cloth. Mispicks are Auc to the running out, or breakage <>\ filling, and the insertion of new filling without removing the par- ticular thread of weft remaining in the shed, and also without in- serting the new filling in the proper shed. In the general line of goods woven, mispicks have not been considered as important defects, hut with other goods, such as napped fabrics and certain classes of multiple harness weaving, mispicks are not allowable. In common loom weaving, they may be obviated by extra pains and extra labor on the part of the operative, who can pick out the particular thread by hand, and turn the loom oxer to Mud the true shed before inserting the full shuttle. With automatic looms, the prevention of mispicks is attained by changing the fil- ling before final exhaustion, so that a full thread is left in every shed. If the filling should break, the loom may he stopped au- tomatically, so that the weaver can find the pick : or. if such breakage is not frequent, the loom may he arranged to run the chance of a mispick at such periods. The mechanism employed for this purpose consists of a simple device called the "Feeler." because it feels of the weft in the shuttle through an opening in the shuttle side, and ahsolutely measures its volume. When reduced to a certain definite quan- tity, the feeler operates to liberate mechanism governing the ac- tion of the filling-changing devices. The cut first printed shows our latest feeler, which is applied at the shipper end of the loom. As shown, it is in contact with the yarn in the shuttle, passing through a hole in the front box plate and a slot in the side of the shuttle itself. Like the Aumann feeler, on the opposite page, it is independent of back lash in lay and position of front plate. The operating parts are shown in full relief, and are few in number. The cuts of the Aumann feeler show the pattern in use just previous. The mechanism at this side end of the loom, however, does not accomplish all that is necessary, for the operation of the filling-changer by a feeler introduces a curious problem, the ejected bobbin having its thread extended through the shuttle 7' AIM ANN FEELER JUST BEFORE OPERATION. This cut shows the form of feeler mechanism devised by Mr. Louis A. Aumann, agent of the Dwight Mfg. Co. at Chic- opee and modified by inventions of W. F. and C. II. Draper. AUMANN FEELER OPERATING. We have had these applied to thousands of looms. They are independent of the wear or alteration in the throw of the lay and therefore require practically no adjustment after the first setting. 7 2 eye to the cloth, while the bobbin itself is in the receptacle, thus leaving an additional thread to he taken care of by cutting apparatus. Unfortunately, this thread does not lie in the path of the regular thread-cutter, requiring an extra cutting device to operate at the proper time. Such a device is attached to the shuttle position detector, which reaches forward to determine the position of the shuttle in the box, as the position detector passes into the path of the thread referred to. This additional cutter not only severs the thread at this point, hut also holds the severed portion taut until the regular thread-cutter severs it again near the selvage of the cloth. While somewhat difficult to describe, the operation is perfectly simple and efficient. The Feeler is practically necessary on certain classes of goods, yet objection has been raised on account of the waste yarn left on the bobbins. We have endeavored to reduce this to small limits by continual perfection of the feeler mechanism itself. We also limit the amount of waste by applying attach- ments to the spinning frames which spin the filling yarn, called "•Bunch Builders," which govern the traverse motion so as to wind a slight preliminary hunch on the bobbins near the lower end of the traverse, so that the feeler will not operate until the bunch itself begins to he reduced in volume. We have patterns of these mechanisms to tit all the American makes of spinning frame. Another objection to the feeler has been raised on account of the extra labor necessary in removing the waste yarn from the bobbins, especially as the bobbins have some- times been damaged by the use of knives for this purpose. We are now building little machines, in which a large, rough- surfaced roller, by rapid revolution, will easily wind off the waste yarn of several bobbins at a time, reducing the expense as compared with the former process, and causing no damage at all to the bobbins. Most of the waste yarn is easily pulled off by tin' fingers. 73 QQ C/3 3 2 H < TO T EMPLE THREAD CUTTER 74 THE DOUBLE FORK. The cloth which we intend to weave on our looms may be roughly divided into three classes. First, including goods <>n which mispicks are not important and on which slight thick and thin places are of little moment. These are produced by the ordinary plain loom of commerce which our regular Northrop loom is replacing. The second class includes the grades on which mispicks are considered important, and for which Ave apply the feeler device. The third class includes all the goods on which mispicks are not important, but on which thick and thin places are not desired. This grade can be woven on our new double-fork loom-- and we expect to find all grades improved by use of this new idea. There is a prevailing notion to the effect that print cloth may have all sorts of faults, because the dyes disguise them. Anyone who looks at the cloth running over the blackboard in the cloth room of any print mill, will notice defects in every single cut of cloth woven, there being full as many with cloth woven on the common loom as with the cloth woven on the Northrop loom. The buyers have grown accustomed to these faults. When it is understood, however, that cuts of print cloth, or any other cloth, can be woven entire, without a thick or a thin place, the trade will undoubtedly demand improvement. The Double Fork system, has already worked with great success on thousands of looms, and has recently been improved and simplified in detail. It detects the absence of fill- ing on either side of the loom, and prevents the take-up from moving if tilling is not present. With two forks, absence of tilling is detected more promptly, anil they also take care of any trouble caused by a dragging end of filling, which sometimes 75 holds the fork up al the left side of the loom, if the yarn is coarse. The double fork is therefore applicable to coarse weaving, as well as very fine goods, having special advantages in each application. It is also added to Feeler looms as an additional precaution. The Filling-Fork, whether single or double, is the most important element of the loom, to our mind. It is liable to false operation if the tines get bent, or if the lay gets out of position, or loose in its bearings. We are now making a fork in which the tines are cast in place in a solid block, and are also bringing out improvements in loom construction intended to prevent the possibility of variation in the position of the lay itself. STANDARD MODELS OF LOOM CON- STRUCTION. Having considered the different new attachments which are peculiarly adapted to automatic weaving, we next show cuts of looms complete with the devices in their relative co-operation with the standard loom organisms w here their detail may he still further elaborated. Although many loom manufacturers have built from one standard set of patterns for years at a time, we have brought out ten different models, with full sets of patterns for each, within a period of ten years. These different models are not only necessary by reason of variety in width and weight ot cloth woven, hut also represent improvements in design of sufficient importance to warrant new construction throughout. A MODEL (ALSO CALLED 1 804 PATTERN ) . \ it now built. This was the loom sent out on the Queen City. Tucapau, and other early orders. We built this model in rights and lefts, not having then adopted our one-hand loom construction. 77 S M \ii ; ll« ■ ifefJiJJF -j« '/ill * ^ -'■ >** PM1 S^ki. mn li^^W^kwi -J § >m ^Wm (A ^ h II i B hi v END VIEW OF A MODEL LOOM. Steel Harness, Saw-tooth Gearing, Shepard Let-off, Mason Take-up, Movable Bobbin-chute and other details as originally presented. "The cloth is as near perfect as can be. Weavers run, or attend, from 16 to 28 Northrop Looms, and do Dot work any harder than I hare Been them do on eight common Looms, and pretty near all the weavers here are what would be called new weavers; thai i-. having only from two to three years 1 experience; and. in fact, the majority "t them learned here." — [Contributor to Wade's Fibn and Fabric. 7S B MODEL (ALSO CALLED 1S9S LOOM). Not now built. This pattern was continually improved and was our standard for prints and other light goods until 1S9S. It had a wider frame than the A model. longer shuttle boxes, new take-up. Stearns rocker and One Hand construction. C MODEL (ALSO CALLED 1896 loom). Not now built. This was our first heavy pattern loom. It was of the One Hand construction with heavy design throughout. (No cut of this to show.) 79 D MODEL, HEAVY STANDARD, NO. 1 HOPPER. Cut shows (l<>l>by head applied. The take-up on this special style of D loom is of the worm gear variety. "Constant progress lias been the watchword of the last quarter <>f a century, and will lead in the next.su near at hand. Mr. Draper puts the Northrop loom, tin- latest production <>t liis model shop, into your mill today and starts it with amazing success, but while this pattern, the product of many years of hard work of the inventor, with the added talents of many mechanics, has I n in course of construction, a new and better way has been devised to accomplish desired results or to overcome Some Slight delect obvious in your lot of looms. And you are told that in the next lot of l< s built these defects will be remedied, and too late you regret that you had not waited before giving your order. The difficulty, however, is inevitable. Evolution is constant in everything to which the mind devotes itseli earnestly, honestly, and persistently — and each lot of looms turned out will naturally be superi- or in some respect to that which preceded it." — \_Pre8t. Frederick E. Clarke at Montreal meeting of the X. E. Cot. Man. Asso., Oct. •">. 1899. So CROSS SECTION OF D MODEL LOOM. No. i hopper, five harness, cotton-harness, Roper warp-stop. "The Xorthrop looms at this mill are running on 60s warp and 70s to 80s filling. I have never seen looms run any better, on coarse num- bers even, than these are running; in fact I do not see how any looms could do better. The weavers run 16 looms each and did not seem to have anything to do. The overseer called my attention to his loom fix- ers on these looms sitting down by their bench sleeping, which he said was no unusual Bight. He says he gets all <>f 95 per cent, product."— [Extract from Expert's Report, June 20, 1903. Si D MODEL LOOM WITH DOBBY. We have sold hundreds of loom-- for dobby weaves which arc giving the best of satisfaction. 82 rpk" FORTY INCH E MODEL LOOM. N<>. 1 Hopper, Steel Harness, original High-roll Cut Motion. ••In conversation with one of our most prominent manufacturers this week, who has just returned from a trip through the South, lie informed us rliat lie took especial pains to visit a mill making print cloths, where it had all Northrop Looms, and that he never saw nicer woven goods, and made at a cost which we are not at liberty to state, but it was very 1<>w indeed.*' — [Boston Journal of Commerce. u i a 84 E MODEL. Regular pattern for prints and sheetings tip to 1904. Cut shows a steel harness bobbin filling loom as made in 1898 and 1899. Improvements have been added continuously 7 , as will be shown in other cuts to follow. "I called at the Mills: tumid the looms running very well. They have reduced the seconds <>n their plain work to 1 \4> per cent, and on their sateens to one-half of one per cent. This is perfectly satisfac- tory to them." — [Salesman's Report, Oct. 24, 1903. S5 E MODEL LOOM WITH FEELER. NO. I HOPPER. The deep can is used to enable the dropping bobbin to drag out the length of filling cut by the extra thread cutter. While the cut shows the feeler on a two-harness loom, it is more customary to use this device on multiple harness weaving. The feeler shown is one of the earlier constructions. "We lnoknl at the Draper loom-. Which are running extremely well, with weavers running 16 loom- each on L-shade cotton flannel, 17s warp and 9s filling. They are doing very well with the feelers and were making little waste." — [Salesman'^ Report of Not 28, 1903. S6 CROSS-SECTION OF E MODEL STEEL HARNESS LOOM, NO. 1 HOPPER. (Shuttle positioning device is different from that in perspec- tive view of E model, and hopper is for cops instead of bobbin. Pulley- are at the left hand on this loom.) This cut gives a good detail of the cloth winding device on our high roll take-up. Also shows hand adjustment of harness jacks. The detail of the warp-stopping connection cannot he shown in this cut. as the devices used are not on the half of the loom which appear- in the cut. S7 E MODEL WITH LARGE HOPPER. This is the regular standard type for general weaving used from 1898 to K104 (still in use). It began to receive the large hopper as per cut in 1 901. More looms have been sold ol this model than any other that we have put out. ••In New England to-day the price <>f weaving on the ordinary looms, wiili the lasl ten per cent, thai has just been given, is nineteen and eight-tenths cents— say twenty cents— per cut, that is, Cor fifty yards. A. new loom has heen invented by which the weaver can mind ahoal twice as many, and therefore the price \»t cut is reduced about one-half. These are what are called 1 1 1 « - Draper looms In the South they have hardly any other kind <>t looms; they have the best. I saw one woman minding twenty-four looms The price they pay for fifty yards in smith Carolina is -ix and one-quarter rents. The operatives "t course, even at this rate are earning more than they ever earned before." .... Georg< Ounton, ss F MODEL. Extra heavy pattern for goods 72 inches and wider. Made with compensating let-off for two beams, triple cranks, com- pound spring- cloth winder, friction pulley drive. G MODEL. Special frame. D Model weight with E Model depth. We have no cuts to show these two latter styles. II MODEL. HEAVY SIDE CAM LOOM. 8 HARNESS. Frame same a- D and E Models. I MODEL, not ready for illustration. This will he of a construction somewhat similar to our present E Model and adapted for the same class of weaving. 8 9 J MODEL LOOM FOR PRINT CLOTH AND LIGHT WEAVES. There are more looms weaving plain two harness goods on print cloth style than on any other single grade of cloth. Mills can equip for this standard product and run continuously for years without necessity for changes. We started originally with a loom for weaving these goods, but in designing foresaw other uses and therefore prepared the frame and other parts for them as well. A year or so ago we made up our minds that there was a sufficient held in light narrow weaving to warrant the building of a special loom primarily adapted lor this use. We have thus developed a model thai take- up no more floor space than is necessary, that is no heavier than is necessary, and in which the moving parts are not clums) and power absorbing. The fatigue of running and handling such a loom must be greatly 9° reduced. We utilize new inventions to reduce shock, jar, smashes, etc., and in view of the light goods to be woven, pro- tect against the slightest crack or thin place by novel mechanisms. Several of these will he particularly referred to in the special articles on take-up. double fork, anti-hang. etc. These various motions have shown so much advantage that we do not intend to limit their adoption to the J Model loom alone. There are several thousand of these looms already running and several thousand more on order. They have proved a great success, especially when fitted with the complete range of devices which we recommend for them. They will run at high speed if neces- sary, and with lighter power. They can he made with either front or back hinders and with either steel or cotton harness, though we recommend the steel harness unreservedly for this class of work. OUR COMMON LOOM. We have at times filled several orders for common looms for parties who were not fully decided as to whether our mech- anisms were applicable to their special kind of goods, with the idea that when we should have the necessary devices they could be attached to the looms. At the present time, however, our range of weaving is so broad that we rarely find a case where the common loom could be advised, and we foresee little future chance for their introduction. Owing to our expensive experimenting and disregard for cost, we probably make the best common loom now in the market. Our common loom is simply our Northrop loom with the hopper and warp stop-motion left off and a slight change at the fork. With our make of loom it is, of course, guaranteed 9 1 OUR COMMON LOOM. The cut shows the common loom of the 1> model type, ol which we have sold several lots to purchasers who bought to equip with Northrop devices later. We have not encouraged the sale of plain looms as our force has been luisv with North- rop Loom orders. It seems strange, however, that tho-e who continue to buy common looms do not universally demand a t\ pe that will he guaranteed to receive future improvements readily. 9 2 that our devices can be easily applied, while this is not always true of looms made by other builders. We have given fully as much attention in late years to per- fecting the conventional loom parts as we have to the betterment of our own additional devices. The common loom which we should furnish would, therefore, have all of our latest improve- ments in the line of let-off, take-up, etc. It is, of course, understood that the cuts which we show do not pretend to illustrate all of our loom products. Each model that we build is made in many widths, and modifications are often necessary. At present, our range in width is from looms for 2S-inch goods, which will, of course, weave narrower, up to looms for cloth 10S inches wide. We call any loom a wide loom which requires additional parts, such as centre swords, double beam. etc. We have found it advisable on these wide looms to use front binders, and a simple rocker motion that will give the shuttle a smooth, straight pick. Some classes of looms require clutch pulleys, which we can supply when ordered, but we do not recommend them for universal application. While we prefer to sell complete looms, we can apply our devices to certain models of old looms of others' manufacture. Such changing over is especially advisable where the common loom is too valuable t<> be discarded, as in the case of broad looms, dobby looms, etc. We have changed oyer several thou- sand common looms with good results and have a special depart- ment for that work. 93 LOOM CONSTRUCTION. Soon after the introduction of our first looms, which were made in rights and lefts, we found that the shuttle used with one type of loom threaded up better than the shuttle on the other, the eyes being entirely different in threading detail. This led to the idea of making looms all one hand, and as this change only necessitated invention in the line of shipping mechanism, we promptly adopted the idea, and have built all our looms in this way ever since. It is, of course, a great convenience to us, as builders, to have all of our looms made from the same patterns, and it must be an even greater advantage to the mills, for not only is their supply of repair parts lessened, but the weavers find it much easier to go through a set of one hand movements, rather, than learn to do many operations with either hand. It seems strange that the original error of complicating parts and detail by right and left construction was prolonged for a full century. It is, of course, still necessary to have the pulleys arranged to belt at either side of the loom, and we find it also more convenient to have our let-offs changeable in position; but the shipper handle is always at the left, and the hopper always at the right, on all looms which we have built with the exception of the A model. When we change over looms of other makes, we supply parts for both right and left hand looms, as no other builder has followed our lead, especially as the system we use is protected by several patents. It is well known that with the ordinary type of loom, as built, one hand will run better than the other, a- patterns of one hand are not precise opposites to the other, and are necessarily better or worse in adaptation. This gives two differently operating constructions to bother the fixer. 94 LET-OFF. Although the Bartlett was our own original let-off, and although we did use it on thousands of our Northrop looms in an improved form, we have now replaced it by a greatly superior mechanism known as the "Draper-Roper," which is self-adjusting and thoroughly efficient for nearly all the possible requirements. We made a curious mechanical error on these motions as first sent out. which tended to give them a bad name, but on discovery of the fault it was promptlv remedied by sending correct parts to every mill where the let-offs were in use, and we now hear nothing but praise for their performances. Like the Bartlett Let-off, the Draper-Roper is actuated from the motion of the lav and governed by the tension of the yarn at the whip-roll. It is, however, additionally controlled by the variation in the diameter of the warp beam, as the warp is woven off, by a follower, pressing against the beam, which by its change in angle determines the limits of motion by which the actuating parts operate. With ordinary let-offs, the cloth woven varies remarkably in width from full to empty beam, whereas with the Draper-Roper this variation is practically eliminated, so far as influence of the let-off itself is concerned. There are other causes which affect the width, and their results should not be confused with the let-off action. A recent test of actual ten- sion at the whip roll during the entire time a beam was weaving off showed that the variation was confined between 33 1-2 and ^2 pounds — certainly a remarkable uniformity for this class of mechanism. "Their Northrop loom* were all running very well: the weavers run 18 prints each, and on the wider looms 16 each: the fixers run 115 Looms each." — [Extract from Expert's Report, • /»/,/. 2. 1904. 95 BARTLETT LET-OFF. The Bartletl was our standard until the Draper-Roper let- off appeared. We owned the original Snell and Bartlett patent and sold over 50.000 of them for use on old and new looms before 1870. 9 6 DRAPER-ROPER LET-OFF AND ANTI-BANG. The cut shows this let-off applied to a J model loom. Note the follower which hears against the warp on the heam. The operative parts are largely hidden from view. Note in the cut of the let-off another new idea which we call the anti-bang. The fro-' slide connects to the whip roll so as to release the warp in case the loom bangs off. This relieves the loom itself from shock and also prevents smashes. We believe this idea will greatly lessen loom repairs and the loosening of nuts and screws. 97 SHEDDING MECHANISM. Our standard forms <>f shedding mechanism at present include the ordinary single roll with strapping at top and cam treadle drive at bottom, for two harness work, with either steel or cotton harness, the Lacev Top Rig for multiple cotton har- m's-, and a spring compensating motion for the top rigging of our multiple steel harness mechanism. We are. however, ex- perimenting with new motions for our steel harness looms, and shall soon introduce a complete novelty in the line of shedding mechanism, doing away with all treadles, cams, and jacks under the warp, giving more space for the warp beam and bringing all of the operating parts out where they are easily observed and adjusted. We shall have more to say publicly about this device when our patents are issued and a further trial made. The Lacey device is simple and durable for cotton harness u-e. and it is always in place to hang a warp, doc- not wear out -trap- so fa-t as the ordinary motion, and is easily adjusted. It is quite similar to the Wvman motion used on Crompton & Knowles looms, but we think it contains important additional improvements; in fact, other loom builders have wished at times to have the privilege of using our motion on their own makes of loom. We are ready to equip looms with side cams for special weaves, or dobbies. when desired. We have built hundreds of side cam looms for corduroy and thousands of dobby looms for various weaves. ••one man who came cinder tuy personal observation was working 27 looms. Be was producing a print cloth, 28 inches wide, 60x64 ends per inch, 29's warp ami 37's weft. The average for the whole mill was about i'.» looms per weaver. I- it possible Cor our manufacturers to compete with thi< "r" — [English expert's report <>n visit to America, from English paper, October, 1902. 9 s DETAIL OF LACEY TOP-RIG ON D MODEL. Our steel harness is becoming so universal that we have less Held for this motion than formerly. Cut also shows our worm -ear take-up with the let-back modification. 99 TAKE-UP AND CUT-MOTION. Although it might have been simpler to stick to standard designs in this line, copying from well known mechanisms, we have, as a matter of fact, given as much time to the Take-Up ol the loom as any other separate feature. We started with a conventional pattern, but on finding that many of our customers desired to weave large rolls of cloth, we tried to design an arrangement which would wind any size roll desired up to r8 inches in diameter. We saw that the High-Roll arrangement of cut-motion seemed to offer marked advantages in this line, although the High-Roll had never gone into noticeable use in this country and was open to many objections in the forms com- monly known abroad. Mr. Northrop devised our present stand- ard construction with the exception of quite recent changes, and the majority of our looms now in use are equipped with the High-Roll pattern. In its best known form, the cloth passes directly to the rough-surfaced roll and is wound around a core, or bar, which is pressed up against the roll by two supports operating from a coiled spring which governs a double gear and rack device. The spring is wound up by the action of the racks as the roll winds, and the cloth is removed by releasing the spring with a hand crank. There are marked advantages in this arrangement, as the cloth will not shrink or wrinkle and the width of the goods will be more uniform and the picks more even. The breast beam comes outside the cloth, protecting it from blemish when the weaver leans over the loom. The direct acting roll also helps take strain oil' the temples and lessens warp breakage. On all our cut-motions we use a metal cloth roll, to which the filleting is applied, unless the goods woven demand some- special surface only applicable to a wooden roll. This will not shrink or swell like a wooden roll, thereby keeping the picks per inch uniform and the yards per pound at a proper standard. We believe the mill that runs wooden rollers will make its cloth either too light or too heavy. If too heavy, the mill is giving away value without remuneration, and if too light, there will be dissatisfaction at the buying end. Quite recently we have made an improvement by which the core or bar in the cloth roll is positively started by having geared teeth engage with gears on the large winding roll when first starting to wind. As the cloth gets larger in diameter on the roll the gear teeth move apart and unlock. While the mechanisms just described are parts of the cut-motion, they are operated directly by the take-up devices proper which transmit movement from the lay or cam-shaft or other moving parts ol" the loom. We have quite a variety of mechanisms for various classes of looms, many of which we have not shown separately, and some of which have been consid- erably modified since the cuts were made. It is practically im- possible to keep our cuts up to date in view of the rapidity of improvement in the devices themselves. "There has been expended in experiments, in investigation and for patent-, some 1300,000. The result is a reduction of one-half in the cost of weaving cotton cloth. The cost of weaving constitutes one-half the cost of labor required to produce cotton doth. Consequently the savins secured by the loom is approximately one-quarter of the labor of producing the cloth. Experts have estimated that in 1895, *80.000.000 was paid for labor in the cotton manufacture in the United States. Assume that the improved loom had been thoroughly introduced, the saving secured thereby would have been approximately *20. 000.000. The interest <>n the national debt of the United state- in' 1892, the last year of Republican control, was $22,893,000. The possible saving of the new loom, therefore, would be about seven-eighths of this interest." — [Hon. Charles Warren Lippitt, ex-Governor of Bhodt Island. B MODEL LOOM TAKE-UP. This take-up derives its motion from the rocking of the lay-sword. It has a let-hack governed from the fork-slide. Cut also illustrates the weft-hammer and shipper knock-off. T MODEL TAKE-UP. Tim take-up is extremely simple, as will be seen by the cut. It is operated by a cam on the lower loom -halt and so timed that it will not take up unles- the shuttle is picked. This prevents the thin places which arc sometimes formed on common and old Northrop looms if the weaver turns the loom over by hand while mending warp or before starting the shipper. The ratchet shaft operates through a worm to the take-up roll — no chance for back lash of gears. A is the upright connecting to the left-hand fork and B the lever connecting to the arresting device. io3 SECTION OF B MODEL LOOM CUT-MOTION WITH FULL TEN-INCH ROLL OF CLOTH. This cut is interesting in comparison with our later motion, which lias mam additional advantages. The fliter or reed- holder shown is not now used. io 4 ORIGINAL HIGH ROLL CUT-MOTION FOR E MODEL LOOM. The cut illustrates our earliest pattern of High Roll cut- motion. It was quickly superseded by the next type shown. io 5 DETAIL OF THE CLOTH WINDING DEVICE OR CUT-MOTION ON OUR HIGH-ROLL TAKE-UP. This is the cut-motion which lias been an integral part of the greater number of Northrop Looms sold. It has been univer- sally satisfactory on the average line of goods. Certain cloth, however, requires greater chance to yield between the fell and the take-up roll, and we have therefore made a new construc- tion shown on the following page, which allows various changes in wind. io6 *l "» fl OUR LATEST ARRANGEMENT OF CUT MOTION. As will be noted in the cross-section of a Northrop loom, as shown in the cut. we have recently made a material modifica- tion in our Cut Motion, in order to cover various requirements of weaving, it being found necessary in certain instances to have a greater length of cloth from the reed to the take-up roll than our former high-roll arrangement allowed. io: This arrangement allows four different systems of controll- ing the cloth between the recti and the roll. The purchaser of the loom can therefore suit himself as to the method employed and adapt the method to the goods. The take-up roll is given a wide range ol* vertical adjustment to allow for lessening the strain on either the top or hottom shade, as desired. The large cut shows a cross-section of the loom without the hopper, in order to emphasize the main feature of the new parts and the three lower cuts show the alternate methods of use. BRAKE MECHANISM. We employ a simple and convenient filling-brake of our own design, which is actuated whenever the shipper is released. We formerly put these hrakes on every loom we made, no matter what the style of weaving. Finding, however, after con- siderable experience, that the action of any braking device is had for the loom in general, we prefer now to apply brakes only to the special weaves where they seem peculiarly necessarv. The illustration on the next page shows the brake attached to the frog in usual manner, also an independent brake actuator liberated by the shipper handle. -V is a rod leading across the loom to operate the belt shipper on the other side of the loom. B shows a detail of the filling-brake lock which is liberated by the weaver before moving the lay by hand. It would be found by close examination, that the filling- brakes on the ordinary looms used in the ordinary mills, are not continuously operative: in fact, it is probable that the great majority do not act as they should. Our own brake ha- the advantage of a positive screw adjustment by which it maybe ioS kept easily adjusted : hut it increases repairs of various kinds enormously to stop looms suddenly, and there is no need of such quick stopping in the ordinary line of weaving. BRAKE MECHANISM USED OX B MODEL LOOM. "Some people say that the Draper loom is apt to make thin stripes, bat fnun all I can hear, thin stripes are about as scarce as hen's teeth. The work runs very well, and .Jesse Barton, an 18 loom weaver, says he ran a loom seven hoars and never stopped, only for dinner hour. It is a common thing Cor looms to run four or five hours at a stretch." — [From letter to Textile Excelsior from Warrenville, 8. C, during 1900. ICK) THE IMPROVED DURKIX THIX PLACE PREVENTER. We applied thousands of these attachments to the old com- mon looms before entering the loom field. Those who wish to get the best results out of their old looms when weaving light goods can use them to great advantage. They lessen thin and thick places, lessen the results of shuttle smashes, lessen warp breakage, and increase production. We recommend them to purchasers of our Northrop Looms who intend to weave light goods on them. Every improvement that tends to lessen the breakage of warp threads is of high importance when endeavor- ing to increase the number of looms per operative. A slight extra cost at the start may pay for itself main times and not always receive due credit for the performance. The construction consists of a pair of arms fastened to the usual bar across the loom which supports or forms the whip roll, and a roller held at its ends by the sliding bearings, noted in the cut by the open hole for the journal. Where Bartlett let- offs are in use the regular roll may be used without necessity for an additional warp roller. In our first patterns there was difficulty at times in adjust- ing the tension of the spring to allow definite control of the movement of the whip roll. We have now overcome this trouble by using uniform spring tension and governing the movement by adjustable stops as shown. We make patterns to fit different styles of looms. I IO SULLIVAN'S PATENT SHUTTLE GUARD. These Shuttle Guards arc made of the best quality coppered wire, five- sixteenths of an inch in diameter, and are long enough to reach the entire length of the hand-rail. An eve is formed in each end, and these eves fit over the bolts which attach the hand- rail to the swords. No other fastening is required, except for certain widths of looms, when a center support is added. The guard fits closelv to the hand-rail lor about three inches at each end and is then bent to hang over the race in any position desired. This form of construction and at- tachment makes the most simple and durable shuttle guard that has thus far been introduced. The hand-rail is not cut or dam- aged in any way in making the attach- ment, nor are there any bolts, screws, or other fastening, such as have to be used with other guards, to work loose and annoy and hinder the weavers. There are no bolt ends projecting back of hand-rail to tear the harness. This guard can be applied for repairs where it would otherwise be necessary to re- new the hand-rail, at less than half the cost of making and fitting a new hand-rail. There are thou- sands of them in use. THE BOLTON LOOM-SEAT. This novel attachment can be applied to any of our looms and is now sent out with all orders, one to each eighl looms. It provides a seat for the operative that is normally held out of the way by a spring. Fig. i shows the seat as held down hv the weaver's weight. Fig. 2 shows it returned to position under control of it- spring. Mr. T. II. Rennie, Superintendent of the Graniteville Mfg. Co., wrote ns he considered these seats an "Indispensable ad- junct to a well regulated weave-room!* THE KEEXE DRAWING-IN FRAME. W e are introducing a drawing-in frame with attachments. especially designed for holding the warp, drop wire detectors, harness, and reed in a new and convenient manner, to assist the operative in drawing in a large number of warp ends in a given time. There has heen some objection to the use of warp-stop- motions in that they caused extra expense for drawing in : but this delect i- largely obviated by this present invention. Its parts are adjustable, and have a range so that they are applicable to all our various forms of warp stop-motions. Price recently reduced one-half. "3 SPECIFICA7FONS OF NORTHROP LOOMS Ordered From the Draper Company, Hopedale, Mass. Make out separate specifications for each style and size of loom. For Date ordered 190 Address Number Si/e Model Right-Hand Belt from Above Left-Hand Belt from Above Right-Hand Belt from Below Left-Hand Belt from Below Kind of Cloth to be woven. Width Sley Number of Picks per inch Number of threads in Warp Number of Warp Yarn Number of Filling Yarn Shall Looms be duplicate of others in the Mill? If so, give date of previous order Is rilling on Bobbins or Cops ? .Total length of Bobbin or C< >ps Note: — It is necessary to send several sample cops with mule spindle, or bobbin and spindle. Our regular sizes of bobbins take ^ 1-2 inch traverse on a bobbin 6 3-4 inches long; 6 i-S inches on a bobbin 7 3-8 inches long; and a 6 3-4 inch traverse on a bobbin 8 inches long. Our regular cop sizes are 5 1-2, 6 i-S and 6 3-4 traverse. Bobbins are patented, and must be ordered through us. At least 200 per loom should be pro- vided. When cops are used we send 30 skewers with each loom lor large battery; 20 skewers with each loom for small battery. These are charged extra. XI 4 Shall we make Bobbin or Cop Heads Standard Butt? Give largest diameter of full filling Bobbin or Cop measured on the Yarn Large or Small Battery? _... Diameter of Spinning Ring Note: — Large Battery takes 25 bobbins or cops. Small bat- tery takes 15 bobbins or cops. What style of Take-up? Note : — Our " High Roll " construction admits of winding any diameter Cloth Roll up to 17 inches. Embodied with this we have three separate styles of Take-up. Our regular pattern takes up with every pick and lets back to prevent thin places. Our Worm Take-up is a positive take-up. without the let-back feature, and is especially designed for corduroys, velvets and similar fabrics, which re- quire 200 picks per inch and above. Our Worm Take-up with let-back is designed for those who require a positive take-up and still de- sire the let-back feature. Our Standard Take-up has 1 1-4 inch up and down adjustment of sand roll. If more is required, please specif v. What style of Let-off? Note: — We furnish Roper. Bartlett. Friction. Roper and Friction, or Bartlett and Friction combined. On " F" Model looms we furnish Compound Let-off: on Corduroy looms we furnish a special Let-off. If Friction Let-off, shall we order Chain. Fibre, or Rope Friction : Will you have Drag Rolls? Note: — These are used onlv for very heavy weaves: heavy denims and goods of this character. We recommend for most cloths Plain Pipe Whip "5 Rolls; for heavy weaves, not taking Drag Rolls, Vibrating Whip Rolls; for very light weaves, Durkin Thick and Thin Place Preventors. Un- less Vibrating Whip Rolls, Thick and Thin Place Preventors or Drag Rolls are specified, we shall furnish with plain Pipe Roll. Will you have Feeler? : What style Warp Stop-Motion is required? NOTE : — We have three styles : Steel harness using one steel heddle for every warp thread, adapted for 2-3-4 ;,!1( ^ 5 harness work. Drop-wire Stop-motion for cotton harness, which requires one drop wire for every two warp threads in a two-harness loom adapted for 2-3-4 an ^ 5 harness work. Single Thread or Lease-rod Stop-motion for cotton harness, using one drop wire for every warp thread. This stop-motion is adapted for any number of harness from 2 up. Drop Wires and Ileddles are extras and should he ordered in sufficient quantities for extra drawing-in sets. It is well to order about 20 per cent, more drop wires or heddles than the looms figure for this purpose. How many Steel Ileddles o]- Drop Wires? How many looms arranged for 2 Harnesses? How many looms arranged for 3 Harnesses? H<>\\ mam up? How many down? How main looms arranged for .4 Harnesses? How many up? How many down? How main' looms arranged for 5 Harnesses?. How many up? How many down? What style Harness Motion? n6 Note: — We furnish the regular Top Harness-motion or Sick- top Compensating Motion. We adapt our looms to take either the Crompton or Stafford Dobby. We also furnish Special Side Cam Motion for Cordu- roys. Are Cams on Cam Shaft or Auxiliary Shaft? If Auxiliary Shalt, shall we send gears to run 2-3-4-:; shade? Single or Douhle Jack Hooks? On what Xo. of Harness shall we set up looms? How many up? How many down? Shall we supply Dobby? How many Harnesses? What style?.... Shall we supply Single or Douhle Spring Jack or Direct Springs ?.... Is Selvage Motion required? Plain or Tape? What Diameter and Face of Driving Pulley? What width of Belt? Tight and Loose or Friction Pulley? Note: — Regular size 12 inches diameter, 2 1-4 inches face, for 2S inch loom. 14 inches diameter, 2 1-4 inches face, for 40 inch loom. We strongly recommend this width of face, as wider pulleys are much more troublesome in shifting belts. For 2 1-2 inch belts and wider, we recommend fric- tion pulleys. We furnish 16 1-2 inch, iS inch and 20 inch Beam Heads. Which do you require? Distance between Heads ? Note: — For proper width between Beam Heads, we recom- mend 4 inches more than size of loom. For those desiring extra space Ave supply Beams g ]-z inches wider than the size of loom. ii 7 We furnish 5 inch and 6 inch diameter Yarn Beams. Which do you require ? Note: — We recommend 6 inch barrel only on tine yarns. How main extra Shuttles? (Only one per loom included without extra cost.) What style Temple will you have. 1 3-4 or 2 1-2 Roll? How many Bobbins shall we order for you? .Style Oil soaked For what number of picks shall we set up looms? Note : — Send us several pieces of reed such as you intend using on these looms. One piece is not sufficient. As the contraction on our High Roll Take-up is con- siderably less on several classes of weaves than on other looms, it would be well to write us before ordering new reeds. The maximum reed space is s; inches wider than the size of the loom. Pickers must be of short pattern, not projecting above shuttle box. We furnish sample sets of strapping and pickers with- out extra charge. On Corduroy looms send us copy of Chain Draft. We will send diagrams of floor plan after questions are answered. By what lines shall we ship? Remarks "The Northrop Loom, by increasing the capacity of the operative 300 per cent., has brought the manufacture i>f cotton up t" a point thai i- considered practically perfect. In its must highly developed form this loom now enables one man to <\<> the work of a thousand men at the beginning of the cotton industry, working by hand.""— From artich on "Evolution of the Cotton Industry," in Gunton'a Magdzim for Feb.. 1904. n8 O M co o c - o w s C a K s o 1— 1 1 =1 1 Proper Width Between Beam Heads 23 S w -* o oc c ?i t •- z c :i t --- 30 C (M ■«* ;•? S bet. © C« **H CO QC O 7! t c 00 O « i< 55 cc cc 55 cc -f -r -r t -r ic ir: ia s 00 a cm ;t ic x Greatest Width of Cloth at Temple * - ^ u; 1- a -< w m 1- c: h co ?( W W X ^ ^ t T t -t -r 1: iff IO l- O) H CO IO l- ir? Da E CO CO ob bin- c o c. -h ?: m 1- h ?; m i- c. -h ^ ct re re ce cc co -f -r -r -r -t ua iq IO NOHCO IO l- iC U3 »C7 — — CEJ CO Total Length of Multiple Steel Harness Space a 2 if IS CCCOWKTT-t^tiniffiCiO CO O ?! -f «E X O lO O CD CC CC CO L- b*v- c o c :t -t c CO O ?! i< 'J GO O W 't CCCOCOCOCOT'T't't'fitliOiO C » O « •♦ !£ 00 ic *o c c e c 'r Total Length of Cotton Harness Space ID ccco^-^-^-ti-jH-^ioihiciicio ca -* so 00 ea •c : c s t- c- .5 ' C en* 00 O Ci -f cc 00 •0 CC CO CC CO — t- Length of Lay ~ for 1.7' , Shuttle 8 Bobbin 3-g 00 O! *t c x a w f a c ?-i L- 00 00 cc x x os -. cc 5: S5 5 ■ •t 1: x ; 01 ^ : CS O O r-l 1-1 t-i 1H bj)s- .E. c en* X - ; ' _ ; ' „ ; ' * £ >R * J' >R >R >R J' >R * >R . ; ' ^ _ ; ' „" — X ^ CM ■* CD 00 O eM^epcOOCvl^epobOCM** Length of Lay foi- 15 1 i Shuttle 7 :, H Bobbin 3-i: T- h in 10 T- os i-t w 10 T- Si >h l- l- CO 00 60 CO 00 Gl 01 Cl 01 ci cc ic T- Si ^ 5r ic .Srg i/j"' U3&0)HCQIQ]>0>HMtOt> Cl i- i- 1- CO 00 00 00 CO Cl Cl 01 Cl 01 - t -t - - t t H ffi IS & O) ri M 90 O --I "-H 7;. ^~ 3-g 1- *^ X O ?1 -t C 00 O W -t ffl X Q L- L- 00 00 00 CO 00 Cl Cl C S C. _ C>! ri- SO X' O CM ri- O O O ~-> —1 ri ■Hfi c/: ^ i> O X O »l ■« C CO O ?H 12 X 1 - 1 - 1 - 00 00 00 oc 00 c: ci c-. c. c. O CM ^ CO 00 O CM u 5 .S"S x co 00 ei •* cc x. 9 nj a qo H ■* CO CO O M J »r: j(t ic ire co co co II 9 Proper Width Between Beam Heads u o c f^ «- i- i~ t- oo oo d if. ~ — ?■ — -- x r ?< -r L- »- I- I- l- 00 X 30 Greatest Width of Cloth at Temple - z SHMIOt-OlHffi -- i- 1- 1- i- i- X 00 ^ [t I- !- 1- b- X X Total Length of Multiple Steel 1 [arness Space u - c 1- I- I- L- X X X 00 Q Q 1) . rz ?! -f -X X O ?! -f i- i- i- i- i- x x x Total Length of Cotton Harness Space go I- I- I- X X X X 00 o i :/ - B C ■?.~ N-*«0000(M-*<0 L- «.- L- L- 00 X 00 00 1 — — 1 w 1 Q Length of Lav for l.V, Shuttle >\' Lav for I.V „ Shuttle 1% Bobbin _o l| -------- i"- =-. — So 7c. i~- ~. — — — ?!?!?!?!?!?? — ^ b : -------- i- ~- T. — T-. 7- 7. — ———?■?!?!?■?■ - j - - 53 '— Z — z ux z -■!/- s =■•- 5 -r 5 = S H S 3 i' S B C 55 — E / - : sogon^oaic u ^ U u ^ ^> X — — X — — O O U 'J '-J ^ _ 11. n o w -b co o = I I I I II I I OOOoOOOOOO-^u :^_: — • — - - 5° ooGoooooou = s CJ U U U '_- u U U U U jj ; J o6o6xxx>xo6o6oox >_^- V V V V U U U i> u u ~* ^ occQWcccsMmmmca g . ~ T ^~~~"T - u - £ abooaoabaocceoii: I — — — ?< ?i ?i ?i ?i — :f oooooooooo*>'3 *3 = ■ r g ~~ - . — ■- x — t > — ~ s. ~ 7>~ ■- x — -• — -^ s — -• — ■J eDfc-fc-fc-fc-fc-ooooooaoaoasaaaaosJSooooOi-i es JJ o : - q, t-fc-fc- t-aoooooooooos osastttS© ooo c = — = - _ £3 - 0« ?'■ ?, V, r; ^, S5z;ns^i^;H£^ ^ ~ ■; * s ?.- c i - - z - - ■ r I -c '— c — x r: ?• — \z.y.~ 7'. — ■- y. ~ -■ — ■-/■-■?< — ■- y ■- - o — — « - > - i - > - i - y. s. s. s. x ~ ~ —. — — 2 ~ - — c u « £ fl* o — ^ o §3 £ = pa §SS u reates Vidth rcioti Temp •-H gs «g O" ow n — 15 — 'j { 5I~ £ u ^"=- ej ?! -t eg u Sa as as -^-■t: 4J u c .__, X " u r pj a| o . ~ — - - - '- _ c — - — — y; :; b- OS r-j t- i- Z e 6t ' r •— - 7~~ '-, —" g ~ c X '_ 1 « = c 111 for Shi Bob S ff 3 C it •' _ 1 '-. — - - -" H : |Z .So ° /, 5 ?: in j INSTRUCTIONS FOR RUNNING NORTHROP LOOMS. The experience of the last nine years is by no means suffi- cient to absolutely settle all points of discussion. We learn more about the art of weaving every week, and consider the possibilities of further knowledge and improvement practically exhaustless. Many volumes have already been written about the detail of plain weaving with common looms, so we shall try to stick more closely to the new features introduced by the novel mechanisms on our own looms. While these new devices necessarily introduce new prob- lems, there is nothing very intricate about their operation. The fact that thousands have been running for years should give the Fixers sell" confidence. HOPPER (OR BATTERY) ADJUST- MENT In setting the Hopper, first see that the filling-fork passes freely through the grate. Then place the filling-motion finger against the filling-fork slide, and the lever on the starting rod at the hopper side of the loom, to which the starting rod spring is connected, can then be set so as to cause the shuttle position detector to clear the shuttle when the lav is at its extreme forward position. Then turn the loom and allow the tilling fork to engage with the filling-motion hook, which will cause the starting rod to turn, and bring the shuttle position detector across the I 23 mouth of the shuttle box. The end of the shuttle position detec- tor should come very close to the back box plate, when the lay is all the way forward. The position of the detector should be 3 15-16 inches from the hopper surface against which the butt of the bobbin is pressed to the inner face of the detector. To see if the detector works properly, pull the shuttle far enough out of the box so that it will strike it. This should cause the latch-finger on the hopper to clear the bunter as the lay conies forward and the detector contacts with the tip of the shuttle. To see if the transferrer acts properly, bring the lay forward with the shuttle in proper position, until the bunter contacts with the latch-finger, and as the transferrer inserts the fresh bobbin, or cop, note how far it is pressed into the shuttle. Should it go too far down and push the bobbin by the shuttle spring centre, the latch-finger must be set further back by means of the adjusting screw at the rear. Should the bobbin, or cop, not go down far enough into the spring to be firmly held, the latch-finger must be set nearer the bunter. In setting the transferrer, it should be regulated so that it will contact very lightly with the bobbin, or cop, which has been placed in the shuttle when the transferrer is at the end of the downward stroke. The wrought iron end of the transferrer, called the trans- fcrrer-fork, which helps to press the bobbin, or cop, into the shuttle, should be directly over the centre of the shuttle opening, and if out of position, should he bent into place. When the shuttle position detector is in proper position and clears the shuttle tip. and the latch finger contacts properly with the bunter, bring the lay slowly forward by hand, and see that the transferrer places the bobbin, or cop, exactly in the centre of the shuttle. If the shuttle should come too far forward or too far hack, the proper position may be secured by tinning the eccentric pins in the lay sword upon which the pitman works. Be careful •-4 and turn both pins, or else the lay will have a complex motion, for one distance between centres will be longer than the other. If the pitman is too badly worn to allow of this adjustment, it should be replaced by a new one. If, by reason of a badly worn picker, the bobbin, or cop. is placed in the shuttle so as to strike high up on the shuttle cover. an additional piece of leather should be put under the leather on the lay end. to compensate for the wear of the picker. The foregoing adjustments will remedy any ordinary trouble. not occasioned by breakage. The hopper, as a rule, gives very little trouble and requires scarcely any adjustment. The rotation of the hopper disc should always bring a bobbin into proper position. The disc bearing should be kept properly oiled, care being taken not to drip oil on the bobbins. If the weavers leave gaps between bobbins when filling the hopper, they may have trouble. They should not allow these gaps to occur, as it is perfectly easy to turn the hopper back and fill it properly. SHUTTLES. The latest Northrop shuttle takes either bobbins or cops. It is shaped to prevent rilling from throwing forward and escaping from the eye. or looping around the horn. As fastened in the wood, there is no chance for catching either filling or warp threads. The spring cover at the rear is inclined so that if the shuttle is too far into the box. the bobbin, when striking the incline, can push the shuttle into place so that the bobbin can enter the spring properly. It the thread entrances to the eye get jammed or closed, they can be opened by knife blade, or other tool, but care should be taken not to open these entrances any wider than they were originally. It" the eye becomes clogged with cotton or lint, it should be cleaned out. A small piece of flannel is placed at the throat of the shuttle for friction, which can he easily renewed. When coarse rilling is used, it may he necessary to put bunches of slasher- waste, or bristles, through holes in the side of the shuttle, to make additional friction. These must he put in by the loom fixers, as we cannot send them out in this way, not knowing just what conditions arise in weaving. If the shuttle spring gets loose, it should he tightened up by turning the fastening screw. Shuttles should not be allowed to run with loose spring's. We believe we have made considerable improvement in this direction by our latest spring and fastening. If trouble is found with cut filling, the wood near the shuttle eye may have become rough, and should be smoothed with line sand paper, or emery. Any small slivers or sharp edges should be removed by the same means. If warp threads should be broken out by the shuttle, it may be that the tips are blunt or rough, in which case the trouble may be remedied by polishing with emery cloth. SHUTTLE BREAKAGE. Outside of the usual splintering and slivering, generally caused by untit wood, the actual breakage of shuttles on Northrop looms is probably <\\\l' to the following causes: 126 The shuttle may get pinched between the temple and the reed, in case the protector fails to act. Our recent models of temples are designed to prevent this from happening. Of course, the fixer should follow up his work and see that the protectors are properly operative. Shuttles have been split by bobbin rings wedging between the spring grips, but this is of rare occurrence. We grind the ends of our springs now, so as to limit the chance of their press- ing against the shuttle sides. Of course, it is possible to break shuttles, if bobbins are caught during transfer, or if certain parts of the loom are broken or inoperative. In spite of all the chances, our shuttles wear very well, considering that one shuttle runs continuously, the wear not being divided between two shuttles, as in the common loom. We furnish all the shuttles used with our looms, so have an actual record of their life, which runs over, rather than under, six months on the average. Excessive wear is often due to sharp reeds. SHUTTLE WOOD. Shuttle wood is liable to curious variations, both from natural and artificial causes. Sometimes the stock is too severely kiln-dried, taking all the life out of the wood so that it breaks like sealing wax. Shuttles are sometimes treated with hot solutions of wax or oil. This may improve the surface smoothness, but if not carefully followed up, may injure the stock. 127 SHUTTLE DESIGN. Shuttles are shaped to run true and balance as well as pos- sible. With the weight continually changing and shifting, as the yarn weaves off, it is impossible to keep the centre of gravity in a uniform position. The shuttle is also pulled out of place by the drag of the vain, which varies in tension as the bobbin or cop winds off. A perfect design would have the shuttle points on a line that would pass through the centre of gravity, with the weight fairhj well distributed on each side of the centre. Shuttles made for front-binder Zooms have a longer back, so that the pressure of the binder in its last contact will not change the direction of the shuttle. We made all our looms with back binders for years, but are now having very good success with front binders on recent models. MIS THREADING. We use this term to illustrate the failure of the shuttle to thread itself properly. With our recent shuttles this fault is almost entirely obviated . It is possible, however, it the filling be weak, or should the shuttle be picked too hard, that the yarn may be broken before it has a chance to thread up. The shuttle eye may possible get jammed or choked by lint so that the thread cannot enter at all. If this happens, the fork will be raised all right, for the thread will draw off the top of the shuttle on its firsl flight. When the shuttle is picked back, however, the thread will be broken, calling for a new transfer of filling and making a curious looking defect in the cloth, as the shuttle uS will continue to lay threads going from the hopper and will lav none on the return. In weaving two shade goods this action puts several threads in one shade. In tact, it may continue this operation until all the bobbins have been transferred out of the hopper. Our present looms are so set as to stop for a double misthread, but even this will not prevent the fault just mentioned, as the fork will be raised intermittently. The misthread detector on the fork will act, however, if no thread is laid in front of the fork twice running. It may be possible for the fixer or the weaver to intentionally disarrange this motion so as to prevent the looms from stopping, but this should not be allowed, as it might cause a bad thin place if the hopper became exhausted or any accident caused repeated misthreading. The fact that the loom is found stopped, even when there is not a warp break or slack thread, does not necessarily mean that the shuttle has been misthreading. It is possible that the shuttle position detector may have prevented the shuttle from receiving a bobbin twice in suc- cession, and this would cause the loom to stop just the same as if it had failed to thread twice running. If the loom is found stopped with an empty bobbin in the shuttle it is a sure sign that the shuttle position detector has found the shuttle out of place. This means that the pick should be set so that the shuttle will go fully into the box or not rebound. Men with inventive capacity often attempt to improve on our shuttle eve, and we do not assume that improvement is not possible where we have made so many changes ourselves. It is necessary, however, to recog- nize the requirements of the case, as a shuttle eye for uni- versal use must be adapted not only tor threading easily, but also prevent the tilling from throwing ahead ami getting out of the slot. It must also provide for easy passage of bunches, be practically self cleaning, give a proper friction, not weaken the wood materially, have sufficient weight to balance the metal parts at the other end, be fitted in the wood so as not to catch 129 warp or filling, and be designed for eas) molding and machine work. As to the simple problem of threading shuttles, as far back as 1894 we could transfer over 1,000 bobbins without a misthread. These records cannot be attained, however, without proper setting of the loom. We believe the set of the pick has more to do with this trouble than anything else, and recommend a light, easv pick with moderate pressure of the binder. We learned years ago that the amount of misthreading was affected by the moisture in the weave room. Yarn is strengthened by moisture and strong yarn will naturally break less under strain whether it is filling or warp. BREAKING OF FILLING. Every break in the filling causes extra labor, as the weaver must put a bobbin in the hopper twice at least in order to have its supply of filling woven off. Every bobbin ought to weave off clean, except on feeler looms, but a harsh pick may break filling by the jerk or cause it to throw" out of the shuttle and catch on other adjacent parts. Sometimes the yarn wraps around the point of the bobbin or skewer while running off. With our earlier shuttle we expected breakage on No. 36 tilling at least one in ten bobbins, whereas we do not now expect more than one in twenty-five. It is easy to note how filling is running by casually glancing at the hoppers in the weave room to see how main partly tilled bobbins have been put back in the hoppers. Filling sometimes catches on the picker OX picker stick. Care should be taken to allow no crack-, projec- tions, or corners where the thread may loop when throwing out of the shuttle. With cop filling the yarn sometimes catches in 130 the slot of the skewer. More trouble is . .. sioned by split cops, due either to shock in the shuttle box or poor design of spindle skewer. This fault can be largely governed by the set of the pick and use of proper checks. There arc many checks in the market which box the shuttle properly, but a shuttle must be received easily to prevent cop splitting, and there are very few- checks which are adapted to this requirement and also to con- trolling the shuttle properly. BOBBINS. We have received a long and varied education in the require- ments of filling bobbins as we have purchased all of those used on our Northrop Looms ever since we commenced to build them. The complaints of our customers therefore all pass through our own office, although up to the present time we have not had any- thing to do with their manufacture. Bobbin wood is liable to ■us fluctuation, especiallv when not carefully selected and carefully dried. We believe the greater part of the trouble with bins getting out of shape is due to short seasoning, it being neces-arv to carry a very large stock of blanks in order to have sufficient supply of thoroughly seasoned wood on hand. Changes in the wood itself not only require reaming and the weeding out of badly warped bobbins, but also cause loosening of the rings re the bobbin- are otherwise worn out. It is. of course, necessary for our loom that the bobbin rings should hold firmly - that the bobbin will lie properly in the shuttle. We insist on careful gauging of both wood and rings at the -tart, but the wood may change after the gauging process. The split rings applied to the bobbins are necessarily somewhat elliptical. In order to 1 3 I obviate trouble from this source the rings are applied so that the slots will not lie opposite each other. The bobbins will swell if filling is dampened so that they will not fit the spindles. This necessitates reaming, hut the reaming should not he done while the bobbins are wet, as too much wood will then he removed. We are now introducing spindles with a cen- trifugal clutch that allows a loose fit with the bobbin on the clutch and allows more leeway for the fit. We believe this is one of the most important improvements ever made in the art. The contour of the bobbin varies with the kind of yarn spun. Bobbins for coarse filling require coarser steps on the cone. With coarse yarn we use \z steps, for print yarn 14. For coarse filling we usually recommend grooves on the barrel instead of ribs. We have made careful experiments in order to determine the proper size of barrel for filling bobbins, and our standard patterns are all of uniform diameter. To avoid trouble with damp filling as much as possible we advise that the bobbins be filled with linseed oil and two coats of shellac applied after they are dried. Much trouble is found with filling yarn because the bobbins do not fit down properly on the spindles. We expect to obviate this trouble entirely with our new spindle, but the fault will necessarily continue in old mills. With the old pattern of spindle the bobbins should tit the sleeve at from one-half to five- eights of an inch, entering the cup (if there be one) at about one-eighth of an inch, fitting loose at the upper bearing, which should be at least 3-4 of an inch in length. Cups are really not necessary on our tilling bobbins as the steel rings prevent splitting. >3- REAMIXG BOBBIXS. When the bobbins are reamed the reamer should be care- fully watched. Not over ^oo bobbins should be reamed without testing the fit. Trv the spindle in the bobbin and feel if there is play at the upper bearing. If not, the reamer need-- spreading. To spread and sharpen a reamer, the temper must be drawn, the reamer placed in a vice and the part that reams slightly spread with a light hammer and a tool made for that purpose. The reamer must then be tempered. Any good mechanic can change the reamer to the proper -ize. A mill with 10.000 filling bob- bins should have at least six top reamers and two "pod" reamers. The upper bearing gives a great deal more trouble than the lower bearing and it is well to have a surplus. Run the reamer at least 2.000 revolutions a minute. — 2.^00 is better. A good man should ream from 7.000 to 10.000 bobbins a day. Every mill should have at least 20 bobbins to a spindle to each number of yarn used. To weave off in the shuttle properly the filling wind should be considered. We have found main mills where changes in the traverse would give better results. On 36 yarn we find best results with the rail going down quick and up slow in the proportion of ij turns on the up-wind to 6 turn- on the down-wind. This is on a traverse of 1 1-2 inches. With coarser yarn like No. 22 we should recommend 1 3-4 inches. PREVEXTIXG BUXCHES IX CLOTH. All weavers know that when the last cn(\ of tilling winds off from a bobbin it b liable to make a bunch in the cloth. Careful investigation has determined that these bunches are 133 practically always due to the bobbins which did not start up properly when doffing and therefore require to be wound on by hand a tew turns in older to piece up. These lew turns are not wound tight enough to wind off properly and very possibly all come off together, which accounts for the fault noted. There is a common method of doffing which also aggravates this difficulty, when the doffers wind the yarn on the bobbins by giving it a few- twists around the base instead of using the socket doff. The socket doff is certainly preferable. In order to avoid the trouble from the bunch with the bobbins not starting properly, Mr. Charles II. Arnold of Grosvenor Dale, Conn., designed a method in which the doffers are provided with bobbins having sufficient yarn spun on them so that they can be pieced up. Whenever an end does not start in doffing, the doffer removes the empty bobbin and replaces it with the bobbin already pro- vided with enough yarn to piece up. In the weaving of hue goods this change reduces the seconds at once to a marked degree. The extra bobbins are of course furnished by spinning a slight amount of yarn on some extra bobbins at the frame and then removing them for use as noted. It is, of course, somewhat difficult to secure co-operation between the two departments, the spinner not often willing to go to extra work on the weaver's account. It is only, however, in this way that good results are obtained. Mi". Arnold's idea is patented, but we allow its free use to all owners of Northrop Looms. '34 WINDING BUNCHES FOR FEELER BOBBINS. The bobbins used on our feeler looms are preferably spun with a preliminary bunch, the object being to reduce waste In- preventing the operation of the feeler until all the yarn and part of the bunch have been exhausted. This bunch is wound about 2 i-S inches from the lower end of the bobbin and is about 3-8 of an inch in length. We supply mechanism especially designed to govern the traverse of the spinning frames to automatically wind this bunch and have them in use in many mills on various makes of frames. They are perfectly satisfactory in every instance where given a little care and oversight. No mechanism will run in a cotton mill without being properly oiled ami cleaned. It is evident that if a feeler loom is set to work with a bunch that every bobbin should have a bunch. Bobbins, there- fore, which fail to start up at the doff should be replaced with special bobbins provided in advance, already having the bunches wound on them. It is. of course, possible to wind hunches on rilling frames without automatic mechanism by simply holding the rail at the transfer point either by hand or by clamp. This method would, however, require special attention by an intelli- gent hand at the proper time. COP LOOMS. In weaving with cop filling more care is necessary than with bobbins. Bobbin filling rarely loops oft. while cops break in two for insignificant reasons. Our skewers are made from i35 conventional patterns by an experienced builder and are designed to lit the sample cops which are sent us. We have to tit the skewers to the cops, as it will not do to assume that all cops are alike because they are spun on similar mule spindles. Some yarn is twisted harder than others and yarn is often spun both coarse and tine on the same spindle. Proper temper is very important, as the skewer should not only have the proper shape, but hold it and stay open. Many Hxers spread skewers with a screw-driver or other tool, hut this is very liable to break them. When a mill uses steamed cops it should he careful to send us sample cops after being steamed. Trouble with cops splitting is not necessarily due to improper shape of skewer or excessive pick at the loom. It may possibly he due to the lack of proper wind in the spinning room. Sometimes cop skewers on our looms get bent by catching in the shuttle. They should he care- fully examined at intervals to see that they are perfectly true. During the transfer the skewer strikes into the box with some- thing of a blow and we recommend that the cop tubes which are removed from the skewers he dropped in the box to make a cushion. WARP STOP-MOTIONS. THE STEEL HARNESS. With our steel harness loarp stop-motion the heddles them- selves are used as detectors to effect the stopping of the loom it a warp thread breaks or becomes too slack. Originally we only applied the steel harness tor two-harness weaving, hut are now using it for two. three, four and five-shade work with great sue- 136 cess. The heddles of the steel harness are suspended by the heddle bars which pass through slots in the upper part of the heddles, the warp threads being drawn through the eyes near the center. The lower ends of the heddles are free from the moving frame, hut are guided by stationary devices which pre- vent their swaying too much either forward or sideways. Be- tween the harnesses is a long, flat casting called the stop-motion girt, which serves two purposes; first, to separate the harnesses and hold them in position, and second, to resist the action of the feeler bar should a heddle drop down and he caught between it and the girt. KNOCK- OFF MECHANISM. Upon the harness earn shaft there is a cam upon which a follower works, which, through a small connecting rod, operates the feeler bars. This cam follower is held against the cam by means of a small coil spring. Between this cam, and forming a part of the same casting, are two projections. Normally, these projections just clear the knock-off, which is a small casting fastened to the same stud or shaft that holds the cam follower. When the heddle drops, the feeler bar strikes it. The cam fol- lower is thus prevented from following the cam, and the knock- off on the shaft with the follower is moved out of its normal position in such a way as to be struck by one of the projections beside the cam, thus moving the whole link on which the cam follower and the knock-off are fastened. This motion of the link is communicated to the shipper handle, throwing off the belt. When a heddle does not drop, the feeler bars oscillate hack and forth, and the knock-off is held out of the way of the projections or lugs on the hub of the oscillator cam, and the loom continues running:. J 37 ADJUSTMENTS. In setting the steel harness stop-motion the first thing to do is to either throw off the belt, or remove the key which holds the end of the shipper-lever in the shipper-handle (in our later looms), and place the shipper handle in the notch in the shipper- lock; this will bring the stop-motion into the same position as when the loom is running. Then turn the loom until the feeler- bars are in their extreme forward position under the girt. The knock-off link should be against its hearing in the hub of the cam, and the cam-follower should bear against the cam in its lowest place. The small casting on the same stud as the cam- follower, called the knock-off, should be so set that it will just clear the projections on the hub of the cam as the cam revolves on the cam-shaft. The cam on this stop-motion is very similar to that used with the cotton harness stop-motion. The position of the oscil- lator-cam is governed entirely by the harness-cams and should work in conjunction with them. When this cam is meshed with the harness-cams, which it does when the harness-cams are on the cam shaft, it must, of course, move with them ; but when the harness-cams are on the auxiliary shaft, care must be used to run the oscillator-cam in the right position. In this case, when the harnesses are level or passing each other, the oscillator cam should be so set that the long axis of the cam is horizon- tally level, or in other words, so that the faces of the cam point directly to the front and back of the loom on a horizontal line with the floor. The cam-follower is held in position In a spring on the stud to which it is fastened; it' it docs not follow the cam as quickly as it should, tighten this spring. Care should he taken, how- ever, not to have too much tension on this spring, hut just enough to make the cam-follower work properly; otherwise the i3» heddle would be bent by the force of the blow. The motion of this cam -follower is communicated to the feeler-bar shaft by means of a connecting rod. the length of which may be varied at will by turning to the right or left. On each side of the stop-motion girt, under the warp and just touching it. are the front rod and back rods, which hold the heddles in place so they will drop into position to be caught by the feeler-bar if a thread breaks. These rods also hold up slack threads which otherwise might allow the heddles to drop low enough to stop the loom. Small castings called heddle-bar collars are placed on the heddle bars to keep the heddles in line with the yarn. There are also guides at each end of the stop-motion girt to keep the bottom parts of the heddles in line. The harnesses are leveled up at the various positions of the crank: On underthrow looms from the bottom center to the front center, and on overthrow looms from the top center to the front center, according to the class of goods to be woven. The harnesses are connected to what are termed harness rolls at the top of the loom. Care should be used to have the back harness connected to the largest roll, and the front harness to the smallest roll, in order to work in harmony with the har- ness cams. In some cases the opposite to this has been done, interfering with the proper working of the loom. The front heddle bars are smaller than the back, and must be set in their proper position. The trout and back rods should be set just high enough to touch the yarn when the yarn is in its proper position on the race-plate. If the shade should be too high above the race-plate it can be lowered by turning down the set screws in the castings at each side of the loom upon which the harness-roll rests, and then tightening the connections between the harness-yoke and '39 treadles by raising the cap with the spring on top and turning it. If the shade should be too low. loosen the connection between the harness-yoke and treadles and raise the harness. The shade should just clear the race-plate. A great advantage with the steel harness is. that after the shade is once set it requires very little or no attention, and new warps can be put in without alter- ing the shade, and more quickly than with any other harness made. In putting in a warp, however, it is possible to get it tangled up; but this can be avoided by a little care and common sense on the part of the operative. After the warp is once placed in the loom there is no danger of tangling. The bottom connection of the front harness should be placed in the second notch in the treadle and the back one in the fourth notch. The heddle-bars must be straight. If the heddles bind in any way on the heddle-bar it will show reedy cloth, and also be a serious strain on the yarn. No oil should be put on the hed- dles or heddle bars. It sometimes becomes necessary to apply a heddle to a harness bar after the warp has been drawn in. and this is usually done by breaking open the eve and slipping it on. While this is all right as a temporary expedient, it is well to go over the harnesses in the drawing-in room before re-drawing, and remove such heddles, as they are liable to catch and interfere, preventing the action of the warp stop-motion. One of the most annoying troubles formerly experienced with our steel harness looms was their liability to become mag- netized, thereby sticking together and making poor sheds. Some slight changes in construction have seemed to overcome this difficulty, as we hear very little Erom it. except on some of our earlier looms. It is perfectly easy to remove this magnetiza- tion by holding the heddles in an electrical coil, and we have demagnetized several lots for our customers. 140 Sometimes the lower ends of the heddles are seriously bent or twisted by the action of the vibrator. This is either due to poor adjustment, which brings a too severe strain, or is some- times caused by improper setting of the knock-off so that a dropped heddle receives several hundred or thousand blows, as the loom does not stop. The same trouble naturally occurs with detector wires as well. Like every other mechanism that contacts with a cotton thread, the heddle is smoothed by use in a way which no previ- ous mechanical method can attempt to duplicate. Our steel heddles will therefore work much better after a few weeks' use, and cause much less warp breakage than when on their first warp. We polish the eyes in the best manner known — in fact we use especially invented processes ; but the rubbing contact of the cotton thread gives the final finish to the surface. It is impossible for this wear to ever make a sharp edge, as the thread turns its corner in such a way as to continually round the edge. So far as our experience goes we see no reason why steel heddles should not last indefinitely. We have had sets running at least eight years that are better than when made. Of course they may get bent or damaged by carelessness, but there is noth- ing in the normal operation to injure them. In our great variety of experiments with various designs of steel harnesses, we have arrived at the conclusion that in order to secure the best results the heddles must be left with absolute freedom to adjust themselves to conditions. Every experiment designed to limit the position of the heddle in any way, for any purpose, has always resulted in excess of warp breakage. With certain weaves it has been noticed that the heddles will not act uniformly, the strain of the shed causing them to sway or bend to excess. Where this becomes serious we have found it advisa- ble to use separators, which keep the heddles from swaying. I 4 I COTTON HARNESS STOP-MOTION, ROPER TYPE. With this attachment, the ordinary twine or cotton harness is used, the stop-motion being applied between the harnesses and the lease rods, two or more threads being drawn through each drop wire. The threads in this stop-motion pass through long slots in the wires instead of round eyes, there being two such slots, — one for the passage of the threads, and the other for the passage of the drop wire bar. We sometimes use a separate free bar or weight passed through the lower slot and resting on the detectors to keep them vertical in action. The feeler bar girt, knock-off, etc., are similar to those already described. We also use a back rod ox warp support, as with the steel harness. The stop- motion girtcanbe raised or lowered and should be set in position for the feeler bar to clear the drop w ires when the shade is wide open and no warp threads broken. It should also lie set high enough so that when the shade is wide open it will not pull the drop wires up to their full limit on the drop wire bar. This can also be adjusted back- ward or forward so as to give room for additional harnesses. The feeler bar, which is the piece of sheet steel bent at right angles with teeth in the edge, should be set so that when it lias reached the end of its forward movement, it will pass under the girt close to it. While this form of stop-motion will apply for many forms of three, four and five harness weaves, there are special classes of shading to which it will not apply. We have therefore introduced the third form, the single thread stop- motion, which can be used with any style of weaving, including dobbies and jacquards. I 4 2 SINGLE THREAD STOP-MOTION. With this construction, there is one detector for each thread. We apply it in several ways, our more common method in the past being to arrange the detectors in two banks, and use them also to do all leasing instead of the ordinary lease nuts. We can make it in three banks if necessary. When used in two banks, there are front and bark box plates instead of the center girt. The feeler bar is different in being a flat piece of steel with notched edges, oscillating between the two banks. To prevent detectors from slipping or bending under the twisting strain, we place serrated pieces of steel on the bottoms of the box plates. The top edges of the box plates serve as warp supports. The feeler bar having double action needs two knock-offs and two connecting rods between the cam and the follower shaft. ADJUSTMENT. In setting this stop-motion, throw off belt or remove key as before, placing the shipper handle in its notch in the shipper lock. Set the knock-off link, (the long casting forming connec- tion to the shipper handle.) against its bearing cm the cam hub so as to have no back lash. Then place the feeler bar in the center between the box plates and adjust the two small castings on the feeler bar shaft which we call the tight and loose oscillator fingers. These should project or hang evenly on each side of the shaft. Now loosen the set screw which holds the stop-motion cam on the cam shaft so as to be able to revolve the stop-motion cam by hand and set the tight knock-off, the small casting fastened to the stud in the knock-off link by a set screw, so that it will clear the point H3 of the cam hub i - 1 6 to i-S of an inch. Turn the cam by hand until the cam follower rests on the lowest point of the cam and the feeler bar is near the hack box plate. Then connect the loose oscillator finger that is on the feeler bar shaft with the cam follower by means of the connecting rod, and adjust the rod so that as the cam revolves the feeler bar will he moved from side to side equallv. When this has been done, connect the tight oscillator finger that is on the feeler bar shaft with the loose knock-off by means of the connecting rod and adjust the rod so that the knock- off will clear the point of the cam huh as the cam revolves. If, when these connections and adjustments are made, the feeler bar should not move an equal distance each side of the shaft, the trouble may he overcome by further adjusting the connecting rods. The spring on the stud which carries the knock-off and cam follower should he set just- tight enough so that the cam follower will follow the cam properly. The tension of the spring on the loose oscillator finger on the feeler bar shaft should be so regu- lated that it will hold the two fingers together on the shaft. RELEASE MOTION. With all of our warp stop-motions except the steel harness. trouble was formerly experienced on account of the feeler bars grasping and holding the detector after the loom had been stopped by a broken end. In such a case the viu\ was drawn in without raising the detector, so that the loom was stopped a sec- ond time, or else the weaver was compelled to find the detector and release it from the grasp of the feeler bar by hand. i 4 4 We now apply with our cotton harness warp stop-motions, devices which automatically release a dropped detector upon stoppage of the loom. This feature involves almost no addi- tional parts, is positive in action, and saves considerable time for the weaver. It is exclusive with us. and fullv covered by patent. SLACK THREADS. Slack threads often cause trouble by letting warp detectors of any pattern drop low enough to engage the vibrator and stop the loom, causing annoyance to the weaver, who may hunt a long time for the supposedly broken thread. Sometimes the trouble is due to the whole warp being woven too slack by improper tension of the let-off. but the greater difficulty is from individual threads. A \ e have tried to arrange sufficient leeway to overcome this trouble, but if it is found serious, the mill should give more attention to its warping and slashing. .Sometimes the relative position of the girt with relation to the whip-roll is the source of the trouble. On some peculiar fancy weaves where many har- nesses are employed, several of the threads will remain neces- sarily slack all the time. If there are hut a few of these threads it is easy to obviate the trouble by letting them run without detectors, as they are not liable to break in any event on account of their slackness. It there i- a great number of loose threads in the pattern, it may be advisable to run them on a separate warp beam . ■45 WARP BREAKAGE. Ever since our first experiments with Northrop Looms, we have continuously run large numbers of them in our own shops with careful supervision and inspection of product, and we feel that we have had more actual tests made of various weaving con- ditions than have been collected by all other experimenters on looms in all time. Some of the results are curious, showing how impossible it is to draw definite conclusions from machinery that employs so variable a material as cotton fibre. We keep an actual record of warp breakage and hud that it varies in different years from as high as 24 warp breaks per loom per cut in one year down to an average of 12 in another, with no perceptible change in conditions other than the quality of the cotton used in making the yarn. All know that the fibre of different crops is not similar. Under the ordinary conditions we expect that the breakage on print warp with either steel or cotton harness should average between 10 and 15 breaks per cut. If warp breakage were to he reduced without attention being paid to other factors, looms would he quite differently designed. In order to produce cover on the cloth the yarn is strained harder in the lower shade and shedding cams are given a jerky motion in order to keep the shades open for the shuttle to pass properly. Our steel harness will break more ends for the first few weeks while the yarn is giving a final polish to the eyes. Bad reeds are liable to cause trouble, in fact man} mills appear to buy their reeds without any consideration of quality whatever. 146 KNOTS. It was figured some years ago that two-thirds of the warp breakage on a loom came from the knots made in piecing the yarn together, as these knots would fray adjoining threads or be caught in the reeds or between the heddles. The number of knots is reduced by spooling from large warp bobbins, and by making good yarn which will have few piecings to cause breakage at the spooler or warper. A certain number of knots is unavoidable, but the way the knot is tied affects the situation materially, in the old hand method the operative at the spooler tied a knot with long ends, so that for some time we advised the tying of a weaver's knot at the spooler, which would not only have short ends, but be less objectionable in size. We believe that in Europe spooler tenders are forced to tie a weaver's knot, and some mills who adopted the practice here found no trouble after getting the help trained, the girls spooling as great a product as before. .Since the introduction of the automatic knot Ivor, however, spooler knots as tied by machinery become much less objection- able as the machine leaves short ends and apparently ties the knot hard and compact. The automatic knot-tyer has gone into such extensive use that our recommendation is practically superfluous. HARNESS CAMS. It is absolutely necessary for good shedding to have the treadle rolls in continuous contact with the cams. If there is too much angle on the cam point there naturally will be more ten- dency to throw. Harness cams should lie set to start opening M7 the shades with the lay at the bottom center oi. the crank. It" tight selvages are desired the cams may be delayed a little, or conversely, for loose selvages, the lay may be pushed back a little. This applies to looms running in the usual American manner, known as the under-throw. With over-throw looms, of course, the setting would be directly opposite. We built several orders of oxer- throw looms for certain of our customers at one time, but found that they had no appreciable advantages which could not be secured as well by simple changes in design on the under-throw principle. As to shape of harness cams we decided after exten- sive tests to use a 6o° rest cam with all widths of loom up to and including 40 inch. If read with relation to the upper shaft, these cams would be known as 1 20° rest cams. On wider looms the rest is made longer until on ioS-inch looms we put on iSc° rest cams. There is no definite fixed rule about the shape of the cam. Different weavers have different ideas as to the amount of rest and the amount of shade opening. We try to satisfy our customers according to the goods woven and the width of loom weaving them. In many cases the proper cam can only be determined after experiment. SELVAGE. Selvage threads are usually looser than the others, often caus- ing the edge of the cloth to crinkle or be longer than the center. This is due to carelessness in setting the temples. If the temple is too far back, the yarn will dra:o around it and stretch the thread, as the width of the cloth in the reed i- greater than in the woven piece. If the temple roll is not free or runs hard for any cause, it will stretch the threads in the same way. Abo it I4S the yarn is not put <>n the yarn beam properly; that is. if it is ■filled higher at the ends than in the center, the yarn will be stretched. Where double threads are used for the selvage and pass through one harness eye, they cannot control the warp stop- motion unless both of them should break at once. Many mills use twisted selvage threads, which, of course, overcome this trouble. As there is more strain on the selvage threads the twisted threads would seem to have an advantage also in lessen- ing warp breakage. CARE OF TEMPLES AND TEMPLE THREAD CUTTERS. To insure proper care of temples, system is necessary and we strongly recommend the practice of all up-to-date mills who have the loom fixers take out the temple rolls and thoroughly clean them and slightly oil the pins that hold the roll in place every time a warp is run out before a new one is allowed to be started. The fixer should also examine the temple thread cutter at the same time. With this amount of care the usual troubles will be entirely eliminated. The temple thread cutter is only supposed to cut the thread leading from the hopper stud to the cloth when the filling is changed. A loose thread at the selvage left by the filling running out will not necessarily be cut by the thread cutter, so that the presence of such threads does not indicate that the thread cutter is not working. These loose threads are com- mon on all looms. In setting temples, place the lay fully forward and adjust the temple head to be about 1-16 of an inch from the reed. The thread cutter knife can be removed by i 4 9 detaching the spring on the cutter arm and pulling the cutter out, at the same time raising the front of it as high as possible. It can be replaced without difficulty. A strip of leather should be placed on the lay opposite the temple heel and cutter arm to strike them when the lay comes forward. The strip at the thread cutter side should he long enough to strike both the temple heel and the cutter arm. FEELER FILLING CHANGER. The/eeler motion is placed on the left hand side of the loom when the hopper is on the right hand side. It is set to pass through slots in the front box plate and shuttle, coming in contact with the yarn on the bobbin or cop as the lay beats forward. When the filling in the shuttle has been nearly woven off so that it will no longer move the feeler, the filling-changing mechanism or />!' Northrop L ns is what we call the hjgh roll. As the name implies, the take-up roll is placed high up. next to and inside the breast beam. This roll has a gearwheel at one end meshing with an intermediate gear which in turn meshes with the change gear, the change gear being driven bv the ratchet take-up wheel, located about half way between the front girt and breast beam. The ratchet wheel is operated by the take-up pawl which is attached to the lay sword, and as the lav swings back, takes up one tooth at every pick. This description refers to the E Model looms. The J Model take-up is quite different. The ratchet wheel is prevented from letting hack by the hold-back pawl fastened to the cloth roll stand. Inside of the hold-hack pawl and on the same stud is the let-back pawl. When the tilling breaks the hold-hack pawl is lifted, allowing the let-hack pawl to let hack the ratchet wheel from one to three teeth, as the quality of the cloth may require, thus avoiding cracks or thin places. The change gear is composed of two gears in one casting, one of which meshes into an intermediate gear and the other into the gear on the huh ot the ratchet wheel. This gear is held in place on a swinging or half circle stand. Each tooth on the large end of the change gear usually represents two picks: for instance, for 64 picks use a 32- tooth change gear, and a 50 gear for 100 picks. After leaving the take-up roll, the cloth is wound on a smooth iron roll called the cloth roll, held in place against the take-up roll by the cloth roll racks. The cloth roll as we now make it has teeth cut in the ends to he turned by gears on the take-up roll shaft, so that the cloth roll will get a positive rotation while starting to wind the cloth. As soon as a little cloth is wound, these teeth will not mesh and the rest of the cloth will he wound by friction alone. *54 The cloth roll racks have teeth meshing into gears at each end of the spring shaft. The spring is wound up by a year and worm wheel 'and handle attached to the front girt. When not in use. the handle can he put in the notch provided for it and be out of the way of the operative. Cloth can he removed from the roll at any time, the weaver taking off cuts when convenient. As the take-up roll is made of metal, it will not change on account of the weather like a wooden one. The fillet is fastened to wooden blocks inserted into holes in the metal roll. The take-up roll is adjustable vertically and can he raised or lowered to adjust the level of the cloth on the lay and give cover. Our new pattern of take-up lets the cloth run over several stationary rolls before giving any contact with the take-up roll, so as to give more Stretch to the cloth between the take-up roll and the lav. which is desirable on certain classes of goods. With the new form of take-up the cloth can he run direct to the roll if desired. The strength of the coil spring on the spring shaft may be varied by turning the collar to which it is fastened. When the take-up roll is empty and the cloth roll is forced up against it. the worm on the spring shaft should be in such a position that the handle by which it is turned should just slide off and'drop into its notch. The loose pawl inside the hold- back pawl has three small holes through it in which to place an extra pin. Each of these hole- represents one tooth on the ratchet wheel, that is. if the extra pin is in the first hole when the loom stops the ratchet will let back one tooth. If in the second hole, two teeth. In the third hole, three teeth, according to the demands of the cloth. When setting the let back pawl, turn the loom over until the filling cam follower or weft hammer i- in its position nearest the breast beam. Pull the filling fork up over the hook on this cam follower and now the change mechanism will lie in operative position. There is a finger fastened to the starting rodhv a set screw which should be turned until it extends under *55 the small arm on the take-up pawl and just lifts it out of its engagement with the ratchet ox pick wheel. This is to accomplish the letting back of the take-up at the time transfer takes place. This should be looked after from time to time with great care. to see that the pawl is actually thrown out of engagement every time there is a transfer, allowing the ratchet wheel to slip around to the extent determined by the pin in the loo>e pawl inside the hold-hack pawl. Otherwise thin places will certainly he caused. FILLING FORK. A filling fork can act improperly by rebounding so as to avoid catching on the hook of the cam follower. Our own fork is designed to balance properly: in fact, we think it the best balanced fork in use. A fork can also operate improperly by being raised by a dragging filling thread, after the rilling in the shuttle is exausted. If the shuttle drags the thread end into the left hand box, unless the filling is rather coarse, it probably will not have strength enough to raise the fork. Our double fork, however, will protect against any such trouble by detecting from the other end of the loom if one fork be operating improperly. If a fork is very light in action, it may be lifted by lint collecting in front of the grid. The more common trouble, however, is due to the lay shifting position, so that the fork tines w ill strike the grid and thus be improperly raised when the Idl- ing is absent. Of course, any false operation of the filling fork when used singly will cause thin places when the Idling runs out. as no change of filling will be called for s, , long as the fork continues to litt. 156 Our double fork gives a double chance against faulty opera- tion : but even with the double fork a shifting lav may operate both improperly. We therefore designed some of our early fork stands to be guided by the lay so that if the loom shifted, the stands would shift also. More recentlv, however, we have adopted a lay guide attached to the loom frame and sliding in another casting bolted to the lav. so that the side position of the lav must always be constant. Filling forks are made in two general stvles, one with soft metal tines, so that the fixer can bend them into any shape desired ; the other made of tempered metal, so they cannot be bent. We prefer to make our forks right at the start, using tempered wire, so that they cannot be bent. In our present construction, the tines are cast into place and their position is absolutely fixed and unchangeable. Our present forks are all made with three tines, although we have furnished four tine forks for special light goods. LOOM LAY. A stiff, heavy lay is absolutely necessary to weave heavy goods, although if the stiffness could be had without the weight. it would probably accomplish the same purpose. The hand rail must, of necessitv. be stiff in proportion. Much trouble is experienced with lays if the wood i> not properly seasoned before use. We find it advisable to rough out our lays and let them season some time before finishing. \\ e carry a large stock of lay timber on hand ahead of orders so that we shall not be forced to use unseasoned stock by any uncommon demand. i57 The position of the pivot from which the lay swings with relation to the position of the crank shaft determines the eccentric- ity of the lay's motion, which is advisable in order to give the shuttle more time in crossing, and also to help give cover to the cloth. After a great deal of experimenting, we have adopted a design suggested by Mr. Robert Burgess, then agent of the Grinnell Corporation, who tested looms of various constructions for us in determining this point. It is, of course, understood that all of these jerky motions make the loom run harder, and probably bring more strain on the warp, but long experience has determined that it is better to sacrifice smooth running to other considerations. The raceway for the shuttle should be absolutely true, and it is advisable to go over looms with a straight edge at times to detect any error. The raceboard should be slightly lower than the level of the shuttle boxes, in order to allow for the thickness of the threads which rest on the race underneath the shuttle. REED. The reed should be either set in an exact plane with the shuttle box bach plates, or slightly back to allow for variations, as it will plane the shuttle if too far front. It should be set at exact right angles with the shuttle race, the hand rail or red-cap being filed to fit, and forced firmly into place. The purpose of the reed is simply to heat in the filling threads, and furnish a back guide for the shuttle. As the dents furnish more or less of an obstruction to any bunches or knots in the yarn, it is advisable to have them a- i«;8 thin as is practicable, in order that they may offer little surface for side contact, and also he free to give slightly when necessary. In order to have a good running- reed, the edges of the dents should be straight and smooth. In nearly every case where mills have complained of shuttles wearing excessively on the hack, it has been because the dents of the reed were sharp, scraping the back- of the shuttle- like a fine hie. and fluting them so that they looked something like a miniature washboard. In the manufacture of reeds, the straightening and polishing of the dents is by far the most expensive and slowest part of reed making, and when not properly done, simply indicates a poor job. and an attempt to make an extra profit. Sharp reeds are also very hard on the warp yarn, the blame of bad running warp often being put on the quality of the vain, when it is really the reeds that make the trouble. To test a sharp reed, draw the finger nail edgewise across it, and if it wears the nail, the reed is sharp and not properly polished. The dents should not bite the nail any. and should, of course, be in line. Manufacturers should insist on having smooth reeds, and inspect them carefully to be sure that they get what thev order. There are reed manu- facturers who supply proper reeds and have pride in their repu- tation. It is not our business to recommend special dealers, but we are often tempted to when noting what inferior supplies are sometimes attached to our looms. The reed dents should be as thin as possible, to allow elas- ticity and can. of course, be made deeper, if the thinning is inexpedient without it. The manner of holding a reed in the lay is not so positive as it might he. since reeds vary so much in contour. We formerly used an adjustable fliter hy which the reed could he positively clamped, no matter what its size. The idea was good in itself, but we found that fixers were liable to screw the bolts up too tight and pull the reed in front of the shuttle box. We have therefore gone hack to the old reed groove system, but '59 have improved its form so that it seems sufficiently efficient. In order to fit this groove properly, it is necessary for customers to send us several pieces of different reeds, so that we may know how much their size varies. THROW OF LAY. When the lav is at the end of its forward stroke it must he in position to allow proper delivery of a fresh bobbin or cop to the shuttle. Any wear of parts that allows the lay to throw for- ward too much should be taken up, and if it becomes necessary to shorten the pitman to take up wear, the position of the lay can still he corrected by adjusting the eccentric pins in the lay swords to which the pitmen are fastened. Of course it is only necessary to adjust the pin at the hopper end of the lay in order to <^et the shuttle box properly under the hopper, hut great pains must he taken to adjust the pin at the other end of the lay exactly the same amount, or else the lay will have a curious eccentric motion, one end beating up further than the other, causing the shuttle to wear into the reed or strike the shuttle box sides improperly. If the wooden parts of the pitmen wear so badly that the eccentric pins will not furnish sufficient adjustment, the wooden parts should he replaced. If too much play is allowed in the pitmen bearings, there is possibility <>f cracks or slight thin places in the cloth when the loom stops. i6o SHUTTLE BOXES. The bach box plates are -et at exact right angles with the lav end plates by filing the ribs or fitting strips at the back of the plates. The back box plates must be set in line with each other. the reed being preferably set slightly back of this line, as it will not do to run any chances of having the reed in front of this line. A long steel stra is necessary in order to trv the plates and see that they keep in position. The front box plates should be set so that the top will lean slightly toward the back box plates, thereby reducing the liability of the shuttle raising in the box. If set at a right angle it will probablv work all right, but it must not lean /ram the back box plate. At the same time it must not lean much toward the back box plate or it will wear the top of the shuttle. With back binder looms, the front plates are adjustable and should be set so as to line the point of the shuttle in the centre of the picker stick slot. With the front box plate in position, adjust the binder properlv bv loosening the nut on the bottom of the lay and the screw which passes through the binder bearing, turning the eccentric bushing with the fingers until adjusted to the proper position. We have had a great deal of experience with different binder materials, at first being ready to follow the request of our customers, until we had definitelv settled the matter to our own satisfaction. A binder may be of wood, wood with leather face, wood with steel face, wrought iron, cast iron, or iron with leather attached. We now prefer a Jen binder faced with leather, as we rind that leather does not wear the shuttle so badly as either wood or iron. Iron binders bring a hard pressure on the shuttle -u^hen the Zoom i with the shuttle part way in the box. the whole force of the momentum of the lav being transferred through the protector rod. binder fingers, and binder to the shuttle, often breaking tss s, as it is i6i pinched in its weakest part. The wooden hinder will give sufficiently to relieve the shuttle, and we think the shuttle boxing is better also as there is more spring to the wood and less weight to be moved. PROTECTOR. The protector mechanism on the Northrop Loom does not differ in principle from that on other looms, so that detailed explanation is unnecessary. On our recent models we use a novel method of adjusting the binder finger, which we think will appeal to fixers. Protector rods sometimes become loose through wear. The caps which hold them can be tightened by filing. The pressure of the binder fingers on the binders is regulated by a protector rod spring in the usual way. Now that we are building front binder looms, we use a novelty of construction which ena- bles us to still employ the ordinary frog and dagger protection. BRAKE. All looms are equipped with brakes, but in one class <>t looms the brake is worked solely from the protector motion when the loom hangs off, while on another class the brake also oper- ates every time the shipper handle is thrown off. The latter system is known as the " Filling- Brake system," for with the com- mon looms the brake is thus applied whenever the loom is 162 stopped by the tilling motion or fork. There is no question but that the application of the brake brings serious jar and strain on a loom. We know this positively, for we have many records taken of looms used with and without the filling-brake attach- ment, showing that looms which do not apply the brake at these frequent intervals, run with much less - for repair, and much less loom fixing. We thought at one time the brake was also responsible for breaking of crank shafts, but further investigation proved that the more frequent reason tor crank shaft break _ came from the strain of a tight belt, as noticed particularly in mills where looms were driven from small pulleys underneath the floor, with short belts necessarily kept very tight. While, therefore, we have a filling-brake sv-tem. ami a m>st efficient one at that, we have recently discontinued its use on looms weaving goods where the pick- were so frequent that the stopping of the loom did not make any possibility of a crack or thin place. On light goods we shall continue to apply them. and the parts, of course, are applicable to looms which may be - : out without them. Our loom has less use for a brake than the common loom as it does not stop for filling exhaustion or break. _ Any brake, to work properly, should be carefully adjusted. When the brake acts by the motion of the frog holder it should not bring pressure upon the wheel before the belt is shipped. The braking surface should be set so a- to bear upon as much of the surface of the wheel as is possible. This can be done by mean- of the adjustment at the bottom end of the brake. The leather on the brake will neeessarilv wear more or less, requir- ing attention in order to obtain the best results. 163 LOOM ADJUSTMENTS. Every new loom will jar screwed parts loose In the first few days it is run. All screws and nuts should be gone over care- fully, tightening them securely when loose. There are main- theories about the proper adjustment of whip-roll, harnesses, and breast-beam or breast-roll. If cover is desired, an extra strain should he brought on the lower shed by raising the whip-roll. breast-beam or breast-roll, or both. Our high-roll looms are pro- vided with liberal adjustment for change in vertical position. Whip-rolls are also adjustable for the same purpose. In weaving drills or twills, strain is frequently brought on the top shade by preference. When this is necessary, the whip- roll and breast-beam should he practically as low as the race of the lay. It is. of course, necessary to adjust the shedding motion and timing of the picks so that the shuttle can pass through the shed without too much friction. These adjustments must vary with the width of the cloth woven, as it is obvious that with a wide loom more time is necessary. Looms are built with the crank shaft set lower than the lay pitman pivot, in order to give more time for the shuttle. The use of a short pitman accom- plishes the same purpose, if the hearing for the pitman is extended, hut this construction necessitates heavier sword castings, and is not so desirable for that reason. The pick should he set so that the shuttle should just begin to move when the lay is in the center of its hack stroke. 164 DRAWIXG-IX WARP. Drawing-in is necessarily expensive, and the question of twisting in warp has therefore been considered. We have made- experiments in this direction, finding there was an actual saving in time of about 1 ^ minutes per warp. The loom was kept from producing, however, during the time of twisting. Of course, warps can be twisted in outside the loom, in a frame made for that purpose. Our steel harness requires no extra labor, while drop wire warp-stops add to the cost of drawing-in. Large beams natu- rally reduce the expense. The Keene drawing-in frame is of great advantage for any of our stop-motions. SIZIXG WARP. Where drop wires are used with cotton harness, it is neces- sary to size the warp with additional care, taking pains to put the sizing into the yarn instead of on the outside, as is the cus- tom in a great many mills. The test of proper sizing is found in the amount of tint noticed, and the average :^arp breakage counted. No. 28 warp yarn should not break more than 10 to 1 2 threads per day with a cotton harness stop-motion on ordi- nary goods. Slow speed at the slasher gives a larger percentage of size. With our steel harness, extra sizing is not necessary: in fact, not advisable, as it may actually increase warp breakage. We recommend the following mixtures for our cotton harness drop-wire system : i6 5 Sizing for Sheetings: ioo gallons of water, 70 lbs. potato starch. 4 to 5 lbs. of tallow, 1 gill turpentine. 1 gill of blue vitriol : boil 20 minutes, or longer if necessary. SlZING for Prints : 120 gallons of water, 60 lbs. potato .starch. 2 lbs. of tallow. 7 lbs. of Victoria zinc: boil from 20 to 30 minutes. Sizing for Medium Weight Goods: 120 gallons of water. 65 lbs. of potato starch. 7 lbs. of tallow. 5 lbs. of alum; boil 30 minute-. For steel harness simply add more water to the above mix- tures. Experiment will determine the proper amount for the conditions presented. LOOM POWER. We believe that all authorities are wrong on the question of the amount of horse-power required for the looms built today. The old experts figured from tests made with light pattern looms, run at low speeds. Every builder puts more weight into his loom today, and higher speeds are in vogue. It is possible that our loom requires slightly more power than the common loom for the same goods, as it uses a heavier shuttle, ami we believe in a stiff, heavy lay. With our first print-cloth loom we had an admirable opportunity for test, as we ran a room of So loom- from a single engine, and could indicate the power abso- lutely. At 190 picks, they showed 33-4 looms to the horse- power, not counting the shafting. 1 66 CLEANING LOOMS. It seems needless to emphasize the necessity of keeping any machine properly cleaned and properly oiled. Different mills have different systems in this respect, some insisting that the weaver shall clean and oil his own looms, while others have special cleaners and oilers. A loom should surely be cleaned and oiled every time a new warp is put in. and it should also be kept reasonably clean hetween such periods. The high-speeded mechanism needs oiling more frequently, and it should he remembered that every place where two metal surfaces are in rubbing contact demands oil. While we have neyer gone into the question of testing oils for looms, we believe that poor oil can do as much harm in the weave room as in the spinning room, and we recommend fol- lowing the advice of competent oil experts, even if their recom- mendation seems to involve slight increase of cost in the oil itself. REPAIRS. It is somewhat difficult to get at average figures of expense in this line, for new looms will need more frequent repair until the weavers and fixers get used to them. We can figure fairly well ourselyes from the amount of parts sold to our customers, although many orders are for parts to he kept as stock on hand. Sometime ago we figured the average repair cost per loom per month at i: iq cents, not including shuttles ox strapping. We 167 understand the repair cosl of the common loom, including shut- tles, is about $3 per loom per year, and we estimate that the cost on our own looms would certainly be under $4: in fact, there are mills using both common and Northrop looms, which inform us that the repairs on their Northrop loom- are actually less than on the common. PRODZCTIOX. Many mills take advantage of the capacity of the Northrop loom for running without the attention of the weaver by start- ing the machinery before the weaver arrives and also running during the noon hour and possibly sometime after the weaver has left at night. In such mills the production is often over 100 per cent, of that possible during regular hours. The compari- son with common looms, which produce less than 90 per cent.. i> interesting. It i- quite common for Northrop looms to give 95 to 97 P er cent - °f product without the gain by running over time. -V mill should not be especially proud of this showing. however, for it simply prove- that their weavers are not spread out over their proper number of looms. It may take many years to kill the popular fallacy that production of cloth per loom is the great end for attainment. Production per weaver is rather the end that should be aimed at. i6S LOOM SPEED. We have never favored high speed for looms, although the Northrop loom can run at high speed if necessary. Simplv as an experiment we have run one of our print looms at 2S0 picks. We have had looms running for weeks at a speed of 220 picks. There is nothing in the addition of our novel mechanism which limits the speed in any way. Our reason for advising low speed, therefore, is not because our loom is handicapped, nor because we wish to sell more looms, as some uncharitable persons have asserted. Increase of speed increases the breakage of warp, requires more fixing and costs more for repairs. Since the introduction of the Northrop loom many mills in this country have speeded their common looms. Perhaps thev wish to wear them out more rapidly and thus be ready earlier for replacement by Northrop looms. We doubt if there is any other good reason for the change. Thev run looms at high speed in England, but simply because of the domination of the trades- unions, which will not allow weavers to run more than four looms. Under such circumstances the manufacturer is bound to get all the product he can from each loom without caring espe- cially whether he increases the number of duties necessarv. COSTS. The common plain loom, as ordinarily built, is largely a foundry product and the cost necessarilv varies with the market prices of raw materials. In 1^94 we learned that an outside 169 builder estimated that a print loom weighing 900 pounds figured $27 for stock, $9 for labor, $3 for painting and $1 1 for general expenses with profit, making a total of $50. Most builders put more iron in their plain looms today, very possibly patterning after our own increase when we first commenced the building of looms. We invite comparison of our loom as a machine prod- uct with any other made, for we not only secure uniformity by machine moulding, but we also put more tool work into the loom parts than any other builder we know. Our foundry castings have a world-wide reputation and our tool equipment for the manufacture of looms is entirely modern. While the prices we charge for our product may seem high, the additional expenses of manufacture must be taken into account, as well as the extra mechanism which we supply. WASTE. We have no very recent figures on this subject. The rilling waste in a Northrop loom print mill, as averaged from several weeks' test, showed .14 lbs. per loom per week on bobbin filling. LOOM EQUIPMENT. The usual common loom, as sold to the trade, includes no extras in the way of parts not secured to the loom, except the beams, 1 1-2 being figured to each loom. Our Northrop loom, I TO on the contrary, is furnished with one shuttle per loom, check stands, shuttle %uard, filling fork, and one loom scat to every eight looms. \\ e also furnish steel heddles or warp stop detectors in quantities as ordered and supply our own temples of whatever pattern desired, at regular prices. The following list specifies the extras which are usually purchased from supply dealers, although we can furnish sample lots, if required, at their prices : Lug straps, lease rods, Jack sticks for cotton harness, strapping, cotton harness, reeds, lease rod holders. We can supply ////';/ place pre- venters on order and also sell extra pick gears, auxiliary shaft with gears for 3. 4. or 5-shade work, selvage motions, etc.. at extra cost. DOUBLE PICK CLOTH. In view of the many attempts at introduction of weaving novelties that produce cloth with two thread- in a shade, we might call attention to the fact that such cloth i> easily woven on our Northrop loom by winding two threads on a bobbin. With this system double production i- assured, but the cloth is not of the regular trade standard. We mention this not to suggest adop- tion, but merely to prevent waste of time on experiment with double bobbin shuttles, needle looms, etc. CLOTH DEFECTS. Cloth as woven is usually inspected for imperfections, such as thick and thin places, cracks, oil stains, scratck-ups, thread runs, wrong draws, too many threads in a harness eye or reed dent. overshots, skips, kinks, loops, unevenness, bareness, reediness, lack of weight, or narrow width. Thick and thin places are usually caused by imperfect action of the let-off or take-up and on the Northrop loom by the rilling fork being out of order. Cracks or slight thin places are caused by the loom stopping and being started, especially if the weaver turns the loom over while mending in warp or placing the shuttle. Our latest take-ups are arranged so that they will not operate unless the shuttle is picked. Excessive looseness of parts in the loom may also cause cracks when stopping or starting. Our double fork will cure thick and thin places and we expect to produce a take-up that will absolutely avoid cracks. Oil stains usually result from carelessness. Care should be taken, for instance, in oiling the hopper stud on a Northrop loom not to let any excess of oil drip on the filling bobbins. If bobbins are allowed to drop on the floor they may get dirty and show streaks in the cloth. Scratch-ups and thread runs are practically obsolete where Northrop looms are used, for the warp stop-motion, if kept in order, will pre- vent either one. Wrong draws and extra threads should be detected by the weaver. Overshots are greatly reduced on our loom, especially with our steel harness motion. Of course, overshots are possible if the harnesses and pick motion are not properly timed. Skips are also caused by improper adjustment of the harness or pick, or if the picker is not in proper position. Kinks result from filling not being properly conditioned and also from weaving goods too narrow for the width of the loom. Also by using a too heavy fork, or not sufficient friction in the shuttle. Too much power in the pick will also cause them. Loops are almost always caused by the harness not shading properly, especially on five-harness goods. Uneven cloth is usually made when the let-off or take-up is not working right, although uneven filling will also give the goods a similar appear- ance. The faults in the surface appearance of the cloth are determined from the standard set by the buyer, and this may vary so that a fault on one class of goods would not be detected on another. Weight and width must be kept right. We believe our Draper- Roper let-off will produce more even goods than any other in the market, and our high-roll take-up principle will also assist in keeping the width uniform. Of course, the weight will vary if the take-up is not absolutely uniform and positive in action. Our iron take-up roll is also of assistance in keeping the picks uniform. Another defect, not always classed as a defect, is the mispick. or lack of thread in a shade or double thread in a shade. With ordinary two- harness weaving the presence or absence of threads is hardly apparent except on close examination. When goods are napped* it is highly important that mispicks should be avoided. In common loom weaving the weaver is personally responsible for a mispick, as he can find the pick by turning the loom over and taking care to make a proper jointure. Some weavers escape mispicks on common looms by stopping the loom just before the filling weaves off in the shuttle. Our feeler mechanism copies this method by changing the filling just before it is woven off. It has been found that the Northrop Loom on three-shade weaving makes less mispicks than the common loom as run in the ordinary manner, for the usual lapse of time between the detec- tion by the fork and the operation at the hopper brings the new thread into the proper shade a good part of the time. The usual weaving expert has more to say about cover on the cloth than any other special feature. Cover is a quality i73 appealing to the eve by evenness and to the feel by softness. Evenness can be positively produced by using reeds having a dent for each thread and may also be apparently produced by weaving with the upper shed slack so that the unevenness is disguised. A soft feel is produced in a similar manner and can also be given by use of soft twisted filling. Cop filling undoubt- edly has advantages over bobbin filling in this respect, although it is possible that bobbin filling may some day be spun with slacker twist ii" desired. Slackness in shed is produced by the relative positions of the breast beam and whip roll, or by the angle of the lay when beating up. Heavy drop wires may take some of the slackness out of the top shed, but we have never found this objection important. Bare cloth is also due to the harness cams not being suitable. Sometimes cloth or warp is soiled by dirt falling through belt holes in the floor above. All mills should be thoroughly equipped with belt hole guards to prevent such difficulty. Sometimes oil from the shafting above the loom will drip on to the cloth or warp. Of course, as cloth is woven from yarn made in other departments, its defects may be due to conditions outside the weave room. If the filling yarn is poorly wound, rings of yarn will slip of, making double filling in the cloth. If not properly moistened it will kink. Yarn may be made from dirty roving or with too much twist. Of course, the slashing of the warp affects the weaving and the goods woven. All the departments of a mill should work harmoniously to produce the necessary result, and the management in charge of all departments is directly responsible for such a result. i74 COTTON MILL PRODUCTS, 1900. (From Census Bulk-tin, No. 215.) ARRANGED IN ORDER OF i'ARDS WOVEN. Total Prints and converters cloths Not liner than Xo. 28 warp Finer than No. 28 warp Sheetings and shirtings Ginghams Napped fabrics Fancy woven fabrics Drills Twills and satteens Ticks, denims and stripes Duck, total Duck, sail Duck, other Upholstery goods Mosquito and other netting Bags and bagging ( !ottonades ( 'orduroy, cotton, velvet and plush Yarns, sewing cotton twine, tape, and other products Total value of all products, in- cluding above 4,509,750,616 1,581,613,827 1,056.278.952 525,834.875 1.212.403,048 278,302.708 268,852,710 237.841.003 237,206,549 235,860,518 171,800.853 12! 1,234.070 11,750,151 117.483.925 50,334,609 41.SS5.023 30,039,616 20.323.1147 7,961.523 «2 1 00 30 *243,218,155 57.780.940 35.010). 575 22.104.3(15 55,513,032 16,179.200 18,231.044 21,060.310 11.S02.794 14.301.302 111.440,033 14,203,008 2,216,371 12,046,637 8,670,384 S75.S0S 2.554.192 2.791.431 2.082,017 a a 450,682 125,000* 100.000 25.01111* 44.227* 45,686 30.000* 28,839* 18,000* 15.000* 5,000* 4.500* 4,421 2.500* S00* 89,588,001 1.709 332,806,156 * Estimated by writer. (The report only separates out the looms on certain lines.) In retelling to the goods which it is now possible to weave on the Northrop looms, it might be simpler to mention those which can not be woven, for the Northrop loom has been suc- cessfully used on the greater majority. We weave all classes of prints, sheetings and shirting, a large line of napped fabrics, drills, twills and satteens, ticks, denims and striped goods; in fact practically the whole held covered by looms that weave with one shuttle, no matter whether they use plain harness '75 motions, (lobbies <>r jacquards. Our Looms have been specially successful on corduroys. They arc also weaving bags, window shade cloth, towels, etc. Quite a number of mills are using our regular loom on goods made with silk warp and cotton filling. We have woven worsted goods by using a wooden skewer to hold the ordinary worsted bobbin. We see no reason why the Northrop principle should be restricted to cotton looms. "We have been running twenty-six of your Northrop looms for a little over a year ami it has occurred to me that yon might lie interested in results obtained. Our percentage of seconds tor the last three months from these looms, for all causes, such as thin places, button hole selvedges, oil cords in filling, etc., is only 2.07 per cent. G Is weigh 2.85 yards to the pound. 18s warp. 15s filling. 1 believe this is a low figure, especially as these goods are all bleached and the bleacbery reports that our grading of tirsr quality is strict so thai they have prac- tically nothing to say to us except to hold the goods up to our standard. Conservative figures show that the looms are producing about '-W.j per cent, oi theoretical production figured on our actual running time. We do not run them over time at all. as some mills do. Some mills ma\ show a larger percentage than we get, bul as the g 1- must hear rigid inspection I think the results produced are fair The looms give US little if any trouble in fixing, and repair account for them is very Light. We are running them I7n picks, which b somewhat higher than you recommend for l~>" reed space looms, bul they give US no trouble in that respect/ 1 — [Letter received from customer Sept. 28, 1900. ••They say they have never hail any complaint from the selling house in regard t<» the quality of their (doth, and some oi the g b they are weaving in 6-cuts rolls, and sending it ou1 even withoul inspecting it at the mill"— [Expert's Tteport of Dec. 12, 1903. 176 PRICES AXD PROFITS. The price demanded for a new machine should bear a per- tinent relation to the profits to be derived from its use. The machine itself may be absolutely efficient, accomplishing all that its promoters claim, and vet demand a price prohibitive by reason of the capital required. On the other hand, a new- machine may be sold so cheaply as to give little encouragement to the builders to continue its improvement, through the only possible channels: namely, expensive experiment. Contrary to a popular fallacy, inventors rarely devote their time and energy entirely for the good of the world at large. Those who develop and introduce the inventions are certainly not so unpractically altruistic. There is no reason why the customer should not pay a proper price for value received : and vet, in the general intro- duction of inventions, it is necessary to give the customer the lion's share of profit, in order to secure his approbation. The value of our spindle improvements has recently been estimated at considerably over one hundred million dollars; and yet the return in price paid for the actual spindles themselves, sold within the period referred to, would be under twenty million dollars, which payment must cover the cost of the spindles themselves, the cost of the patents, the cost of expensive litiga- tion, and all the experiments, advertising, and general expense connected with the industry. The introduction of the spindle was comparatively easy compared with the introduction of the loom, for the early price of new spinning with high speed spindles was actually less for a given product than the slow- running frame-, while with our loom the price is nearly three times the price of the competing loom, so far as the amount of product is concerned. There is always 77 a protest against higher prices, no matter what the advantages may be. Looking at the introducer's side, it is evident that, having hut seventeen years of patent protection, several years of which are usually used up before actual sales are made, he must make enough out of this limited period to repay all of his expenditure involved in perfecting, protecting, and introducing his idea, as well as a fair bonus to repay for the risk of attempting to improve in the first place. The profits must also cover the expense of hundreds of useless experiments, thousands of dis- used patterns, possible litigation, extensive advertising, replace- ment by improved parts, etc. It may be easily demonstrated that if it had been possible to sell all the possible customers all the looms they could use at a uniform price, none of them would derive appreciable profit from the operation ; for the competi- tion amongst themselves would reduce the profits till the general public received all the advantages of the new economies. The earlier purchasers of our looms would, therefore, pre- fer to see our introduction gradual, and it would hardly be fair to them to reduce prices in favor of those who were not so willing to assist by patronage in the early years of trial. We have no doubt but that we could have sold a great main more looms, had we set our price lower in the first place. We might even have made as much profit; perhaps even more. It would have been necessary, however, to have still further enlarged our plant for such a purpose, and after filling the more numerous orders given to replace old machinery, we might easil) have found ourselves over equipped for the regular business of supplying new mills for the future. The possible profits of the Northrop loom are based on the actual fact that with them a weaver can produce at least twice as much cloth as formerly, often three times as much, and in special instances even more, by tending a much greater number of i78 loom:-. It is also found that the Northrop looms will produce more cloth per loom, as they generally run for a greater percentage of the time and in many mills are allowed to gain still more by run- ning during the noon hour. The quality of the cloth is often bet- ter for certain purposes, but we do not claim yet that the improve- ment in quality actually increases the price at which the cloth can be sold. We do believe it is enough better to give a preference and we believe that with certain of our later devices, employed in large quantity, we shall actually create a new and better grade of cloth which the common loom does not produce. The weavers on Northrop looms, bav- ins: actually less work to do. even while tending: three times as many looms as formerly, have been allowed to share somewhat in the profits bv being allowed a price per cut at which they can make better wages. The average piece price for goods woven on Northrop looms is probably a little less than half the former weaving rate. To offset this gain we have an increased cost of the loom itself, with loss of interest on the extra investment money, and a very slight increase in repairs and fixing, although there are mills which claim that their expenses in this line are actually le>s with the Northrop loom. Roughlv figured, the gross profit on the loom should run from $20 per year per loom upward. It varies with the scale of wages paid, and the number of common looms formerly tended: for instance. Northrop loom weavers are paid six cents per cut in Southern mills on goods where they might earn nine cent?- in the North. The weaver that changes from four common looms to twelve Northrop will show a greater gain than one who changes from eight to twenty. There are many incidental advantages in the lessening of the number of operatives required. When we take half the help out of the main department of a mill we greatly lessen the num- ber of tenements necessary, lessen the cost of bookkeeping and paying off. and less personal attention is required from the i79 supervisors. Our loom being automatic in character, requires much less skill and training from the operative, for it is easy to learn to run Northrop looms; in fact, green help become accom- plished weavers in a much shorter period than with common looms. As the loom is automatic and therefore more responsi- ble for errors, there is less chance for trouble with the weavers over bad work and fines. Some of these matters may seem small in themselves, but thev amount to considerable in the aggregate. We have labored very hard to overcome traditions in weav- ing that have grown up out of the long ascendencvof the common loom, and we believe that the possibilities of automatic weaving are still hampered by customs originating with common loom practice. When a weaver was limited to four, six or eight looms, it was more or less a matter of pride to keep them run- ning, and if the weaver could not keep a certain number con- tinuously operating he was forced to use a less number. This bred the instinctive horror of a stopped loom, which prevails now that the Northrop loom allows a much greater number to the operator; yet economv actually demands that a weaver with automatic looms should have enough under his charge so that some stopped looms would be more or less of a necessity. It is quite common in Northrop loom weaving to have production run as high as 95 per cent, of the possible production without counting in the extra gain by running noon hours. It is a com- mon thing to see a Northrop loom weaver with all of the hop- pers lull and no single loom stopped for any purpose. Such a state of affairs simply proves that the same weaver could be given a greater number of looms if it would be possible to educate him into a state of mind that would not look on the stopping of several looms at a time as a terrible error. It can be easily proved that it would be much more economical for weav- ers to get So per cent, off of 30 looms rather than 00 per cent, off i So of 20 looms, or 95 per cent, off of 16 looms, provided the pay of the weaver were regulated to the product in proper propor- tion. We believe it for the best interests of the loom, the help and the management as well, for the Northrop loom weavers to be relieved of the work of cleaning and oiling their looms. No labor-saving device attains its full efficiency in the first few years of use. Our later large hopper looms have certainly enlarged the scope of the weaver, and continual improvement will gradually reduce warp breakage and other loom stops due to various other causes. The problem of how to increase earnings is often solved by enlarging the plant, but less money applied to the improvement of a present plant may sometimes give far greater returns with much less inconvenience. The change from common to North- rop looms requires no addition to floor space. As above noted, it greatly decreases the number of operatives, and therefore solves a most perplexing problem in localities where weavers are scarce. If the old mills will not appreciate these facts thev must face the competition of the new mills, which start with more modern equipment. We are frank to say that the hesitation of many of the older mills has been distinctly disappointing, for we should like to see them share in the benefits of our new ideas on account of the friend- ship founded on long and intimate associations. Failing to induce them to take the majority of our products, however, we must in justice to ourselves encourage the building of new plants. We should, if necessary, place our looms, even if we had to build and operate mills ourselves in which thev were used: for we are absolutely convinced that the mills with our machinery can make profits in straight competitive lines at prices which will drive the older, poorly equipped mills, out of busi- ness. If there is demand enough to make a profit for all, the mills with our machinery will make the greater part of it: and iSi when there is no profit at all for the older mill, the newer mills can at least keep a balance on the right side of the ledger. According to the census reports there were in 1900 about 450,000 cotton looms running in this country alone. In 1904 there are certainly over 500,000. Out of this number there are probably at least 75.000 looms running on tapes or narrow wares and with box motions or other devices that practically take them out of the field of filling changing mechanisms. These looms, however, offer an opportunity for warp stop-motions which we have already accepted to a considerable extent. Taking out the Northrop looms already delivered and running, there remains a field of about 330,000 looms for us to replace, as this number of common looms is still used on goods which we are perfectly capable of weaving. With our present plant, even before recent additions, we attained an output of 2,000 looms per month. With our new foundry facilities and a proper increase in tools for which we have space already saved, we could undoubtedly deliver 40.000 looms a year. In view of the looms sold to new mills it is therefore somewhat doubtful as to whether we could entirely replace the old looms in 10 years' time, espe- cially as we should be foolish to increase our capacity to an extent not warranted by the normal future demand after the old looms are replaced. The trade can therefore be assured that those who have purchased looms now will have at least 10 years' advantage over those who delay. The earlier purchasers of our looms have long since paid for them by their profits, and these profits are practically guaranteed so long as there remains any appreciable number of common looms in use. During the last few years the trade has noticed main peri- ods of curtailment by large numbers of mills running on certain standard lines of goods. It has also been noticed that other mills on these lines of goods have not only run full time, but even kept running during the night hours in spite of the disad- lS2 vantages of such a practice. The main difference be- tween these mills has been that one class run common looms and the other Northrop looms. It is not to be supposed that the introduction of a revolu- tionary machine like the Xorthrop loom is effected without diffi- culty, annoyance and delay. Those who use common looms and haye not immediate chance for replacing them are naturally anxious that their competitors should not adopt advantageous improyements. Those who sell common looms are adyerse to acknowledge the merits of their competitors and the influence of a large body of manufacturers with their salesmen and per- sonal friends is of acknowledged weight and importance. There is also a limited class who haye made unsuccessful experiments with certain lines of weaving with the new deyices and who are not disposed to admit that the other mills can be more successful than themselves. All of these opposing elements together create a certain atmosphere of doubt and a disinclination to accept facts. which can only react to their own disadvantage. Apart from the profit derived from the sale of our looms there is a distinct personal satisfaction in overcoming the antagonism of these varied elements and proving the truth of our earliest contentions. It has always been held to be a difficult matter to convince a man against his will, but difficulties in the undertaking make success so much the sweeter. Many have read the series of letters that were written to the Manchester Guardian by their special correspondent who visited this country with the delegation that inspected our cotton industry. Nothing recently published gives an equally clear and compre- hensive view of the trade situation from North to South by an outside, and therefore unprejudiced, party. The following quotation is but one of many which refer to the paramount advantages of our loom : "The mill contains, at present, 25,000 ring spindles and 800 Nor- throp looms. All the cloth manufactured i- for export, and consists of i»3 two kinds only, namely — China drills and sheetings or shirtings. Drills are .'fo inches wide weigh •'< yards to the pound, and have 68 ends and is picks to the inch. The sheetings are 36 inches wide, are of the same weight as the drills, and have 48 ends and 4s pick- to the inch. In both cases the yarns are 13.65s twist and 13.80s weft, the CUtS are 120 yard- long, and the piece rate for weaving is 13 cents a cut. The rate for weaving similar drills in Maine. I had found but a tew days before, to lie 58 1-2 cents for 120 yards, and that was less than the Lancashire rate. Mere, the cheapness of the Southern labor and the use of the Northrop loom had enabled the superintendent to undercut the Maine weaving price by 75 per cent. One man who was running 24 looms told me that lie could earn $1.35 per day: two other men were also running 24 looms each, and said they could make 81.50, . . . the tacklers tend 100 looms each." The writer also refers to a statement made to him in Massa- chusetts to the effect that the Northrop loom is so easily managed that an inexperienced girl learned to run 14 of them within a week. It is not often that a manufacturer will personally admit the extent of his profits by use of the Northrop loom. Recently, however, it became necessary for such a manufacturer to file an affidavit, which, being a matter of public record, we quote in part, although witholding the name for the present. In referring to a large number of looms running with Northrop attachments, the affidavit states as follows : '• This mill is one of the most modern iu this country so far as equip- ment is concerned. The average pay id' the weaver- who attend to these looms (common) that weave such goods is nine dollars a week. Each weaver takes care of four looms. The average production of each of these looms is twenty-lour yards or twelve pounds of such goods per day. This would he one hundred and forty-four yard- or seventy-two pounds per loom a week, making five hundred and seventy-six yards or two hundred and eighty-eight pounds of such goods a week for the four looms taken care of by each weaver. This is the only mill of which I have knowledge where the weaver can take care on an average of as many as four looms."* (On this Style of goods. "The cost of manufacture id' such goods for the wage- of the weaver only is about 3.12 cents per pound. With less improved looms for producing such g 1<. of which many are in use. the cost is greater as a weave]- cannot take care * • t so many looms.'' The affidavit then states that the use of our devices on these goods increases the production to 38 yards per loom, or 10 pounds of such goods a dav. As a weaver attends six looms of the new 1S4 style, the production per day per weaver is 228 yards, or 114 pounds of such goods. The cost per pound is about 1.31 cents. or a saving per loom per year of over $100 each. The affidavit states that the profits from such looms will be about 9 per cent. on the entire cost of the plant, including carding and spinning machinery, and if the plant were to consist solely of looms, the saving would pay a dividend of about 19 per cent, on the cost. The affidavit also calls attention to the greater product per loom as requiring less looms, less floor space, etc. In fact 100 looms at this ratio of product would do the work of i^S common looms. On this basis the saving in number of looms and floor space would possibly pay for the entire cost of the attachments, as these are one of the most expensive tvpe of loom built. Of course, it is evident that this is a pecular class of weav- ing, inasmuch as the weaver only changes from four looms to six ; yet the greatly increased product shows that the weavers on six looms are producing more than twice as much cloth per weaver compared with the common loom product. This affidavit was not made with the intent of aiding us in any way by its information; in fact, we only ran across it by accident. We recently learned from an Indian cotton manufacturer, now in this country, that in India his weavers run two looms each and earn $7.50 per month. This seems a very low price, but as a matter of fact it is $3.25 per loom per month, or $39 per loom per year. There are plenty of Northrop loom mills in the United States where the wages are under $20 per loom per year, although the American weaver may be earning five times as much money. Of course, it is probable that Northrop looms may invade India itself and the coolie may run four, or eight, or sixteen, instead of two common looms. Theoretically, all manu- facturing could be done cheaper in such countries as China and India — but practically the high wage countries hold their own. iS 5 Yet the only reason they do hold their own is because they take prompt advantage of economical methods and devices. The mills that defer using Northrop looms until India is equipped, will have to face a serious proposition. But why should they wait? We were recently permitted to see a record from the hooks of a large Northern mill using both Northrop and common looms. The figures were based on a low scale of weaving wages for the common loom. The figures showed an actual difference of $23.52 per loom per year in favor of the Northrop loom above all extra expense for supplies, fixing, cleaning, etc. The weavers on the Northrop loom also earned $55.12 each, per year, above the earnings of the common loom weavers. This record is based on sixteen Northrop looms to the weaver. Some mills already run twenty-six Northrop looms to the weaver. Recently noting a broker's list of Southern cotton mill stocks for sale, with prices bid and asked, the writer, as a matter of curiosity, separated out the mills which had bought Northrop looms, and figured a comparison in the value of the stock as quoted. The price asked was taken in each case, the price bid being added in only where there was no asking price. The total result showed that 28 mills without Northrop looms averaged a stock value, as thus figured, of $102 a share. The 37 other mills, having Northrop looms, averaged on the same basis, $114 per share. Prices of looms vary somewhat with cost of materials and equipment desired. Thev should properly vary in proportion to the expense and utility of new attachments. We do not, how- ever, add to the price of our loom when improving its funda- mental features. It has been estimated that we have actually added $15 of cost per loom to our complete machine since its earlier stages. We are sflad to estimate on whatever looms iS6 are desired and specified. Old common looms arc taken in exchange at fair allowance under certain conditions. Our policy of smashing up old common looms taken as part payment for new Northrop looms has awakened a certain amount of comment, the visiting Englishmen being particularly im- pressed. Of course, some of these old looms have outrun their utility and are fit only for junk in any event. Many looms thus replaced, however, have been comparatively new and certainly efficient so far as common looms may be efficient. One of the frequent English visitors to our country pub- lished a comparative criticism of the Northrop loom on his return home, that endeavored to show how little actual saving was possible. In view of the wide circulation of the article, as copied by various trade journals, we thought best to issue an answer at some length, taking up the various comparisons in detail and explaining the falsities on which the final figures were based. We were rather embarrassed in replying by the fact that while the Northrop loom mill was well known to all, the common loom mill selected by comparison was not named, and the assertions of speeds, wages, etc., relating to that mill, could not be verified. Without repeating our argument, we might say that we found several reasons to criticise the assumptions made, and if any expert who cares to venture further in this line will give us detailed information as to the source of his facts, we will be glad to enter into a further discussion. The comparison of one mill in one definite locality, with another mill several hundred miles away, is not necessarily convincing. The best comparison possible is that of Northrop looms and common looms running in the same mill, under the same conditions. Our best customers include the mills that have made this experi- ment for themselves, and we are ready to contend that these mills are perfectly capable of figuring cost and appreciating conditions. iS 7 "How the introduction of this new loom affects the cost of labor may be shown by a comparison of two accounts of the cost of labor in print Cloth, line taken by myself from a mill account of older date, hut from line of the best mills in New England, and the other fr the workings of recent date, received from a mill hut a few days ago. COST <)F LABOR IN ONE POIND OF PRINT CLOTH. (28 inches, 64x64. seven yards to the pound. Carding Spinning Preparing for loom Weaving < Hher lahor expenses Total lahor cost Difference on account of improved loom All other differences lss7. Cents. 1898. ( ents. 0.855 1.137 0.697 2.S 0.239 0.7 1.1 0.7 1.6 0.25 5. 728 4.35 Differences. 1808. Cent-. 0.155 0.037 -0.003 1.2 -11.011 1.378 1.2 0.178 The items covering all other manufacturing processes are scarcely worth noticing. The difference is almost entirely traceable to the new loom Now, by no possibility can the strain which the Xorth could he subjected to by the South be so great as the strain the Northern mill has to sustain from Northern mill, and the Southern mill from South- ern mill ; for the same causes may be found in operation in the Smith that produce the differences in the North. The differences of this pro- nounced type are created by the introduction of the so-called "auto- matic" loom. When, by this change, 50 per cent, in the weaving-cosf can be saved, it is obvious that it will not take long to convince mill- owners that it is profitable to discard the loom which was satisfactory until very recently, and to adopt the new loom by which an expert weaver can turn out from two to three times as much cloth in a week." — [Jacob Schot nhof. ••The manufacturers are perfectly willing to try any new device that may come out in the way of new machinery; and no better exam- ple of that can he given than the fact that the automatic loom ha- round its home in the Smith almost exclusively, and the advantages of the automatic loom are. by the Southern manufacturers, deemed t" lie very much in its favor, as against the ordinary running loom. Several man- ufacturers said, in fact, that it was QO i hard to secure '.'7'o per cent. <'f the full possible production of the loom." — [Mr. .u> /■-•< /•. A'. )". Journal of ' 'ommerce. INS "GOOD WEAVING WORK.— A correspondent at Spartanburg, s. c. writes us that they have weavers at Spartan Mill No. 2 running 30 Draper looms. One is a woman, and she has taken off iu February up to the night of the 13th, 326 cuts, 51 yards to cut, which is 50 35-100 yards per loom: speed of loom 180, 64x<'>4 goods, which makes 97 86-100 of production. How is that for running Northrop Draper looms?" — [Textile Excelsior, Feb. 18. 1899. ••There can he no doubt that the enormous expansion of the Ameri- can cotton industry during recent years has been very largely owing to the Northrop loom, and the conviction is steadily gaining ground in this country that only by the general adoption of the Northrop loom can our cotton trade lie put once more upon a thoroughly sound basis."— [Letter from London correspondent to The Indian Textile Journal, printed September, 1903. One of the cloths made here very largely in the 40-inch looms is 32 inches wide and has 6S ends and 112 picks to the inch of 42*s twist and 36's weft. It is woven in 62 yard cuts, and the price paid to the weavers is 27 1 4 cents per cut for the Northrop loom and 56 cents per cut for the ordinary loom. The latter is, I believe, 10 per cent, less than the rate paid in Lancashire, but the ordinary eight loom weaver here can earn *9 a week and the Aveaver who runs twenty Northrop looms £10.50 to £11. — [Correspondent of Manchester Guardian. For the 2000 Northrop looms there are 134 weavers — a number which I verified by counting the names in the overseer's wase-book. Some of the weavers are running 20 40-in. Northrop looms each, others 16. and a number of learners have 12 each, the average for the whole of the 2000 looms being a fraction less than 15. . . . — [Correspondent of Manchesti r Guardian. ••< ailed at the Mills : found them exceedingly pleased with the Northrop looms. They are getting an average of between 26 and 27 yards per day. which is more than two yards more than they get from their common looms. They are weaving 78x80 goods. 40" wide. 52 yard-, and pay 20 cents a cut against 42 cents. The weavers are run- ning 20 looms: there are two lixers on 204 looms, and the only extra help in the room is two boys for cleaning and oiling." — [Salesman's Report, Oct. 10. 1903. ••Their weaving is running extremely well, and they have on 1182 Looms, which they have been running an average of about 19 1 ., looms per weave]-, and Mr. is sure they will lie able to bring it down to an average of 22 looms to the weaver throughout." — [Expert's Report • Y o. 14. 1903. 1 89 "Mr. said the only fault he can find with the Northrop Looms today is that they use too much filling. Since he came here he had had to pur two extra spinning frames on to spinning tilling for these loom-, and now he has just put on the third.'* (In another mill . "Mr. .the overseer of weaving, say.-; they are getting 93 per cent, product from the Northrop looms. 26 looms to a weaver. 163 picks per minute." — [Extract from Expert's Report, Dec. 12, 1003. ••The work at this mill is running very nicely indeed. They now have some weavers running 30 looms each, and with all their looms running — 1292 I understand — they have only 59 weavers at the present time, and expect to spread the weavers further the coming week." — [From Expert's Report of J shape is novel, as manv mills use similar forms. We find that the cop waste on the new pattern is less than one-quarter of that with the regular skewer. We have to pay a slight advance for this extra process, but shall not make any difference in the price charged to customers. We have also improved our skewers by adding a flat band to the head below the rings to prolong wear. Referring to other improvements we might mention that our cop hopper now has a neat device which prevents the rota- tion of the lower cop skewer in the hopper — a fault which some- times causes the thread to wind around the skewer and break when the transfer of H llin^f is made. 2l6 Our shuttle-positioning device has been greatly simplified, and now operates with less strain on the fork and slide. We are introducing a new temple thread-cutter, in which the old trouble of wear on the roll pin is entirely obviated, for the roll pin is no longer within the path of movement of the cutter, even after wear has occurred. The temple itself has been improved considerably in detail in other lines. Our shuttle, with the Jonas Northrop eve, has had so rapid a success as to seriously embarass our manufacturing depart- ment for awhile. We find this eye saves materially in prevent- ing the breakage of filling, both during the transfer of filling, and later while the filling is running out. We have made slight changes that still further lessen the old percentage of faults. In our warp stop-motions, we have introduced changes in position and arrangement of parts, which are trivial in detail, but accomplish remarkable results. We have print looms in our testing room that run for weeks with an average of about six broken warp ends per day. We have discovered new -methods of manufacture which will make our steel heddles start up with less breakage than formerly. We have always told cus- tomers that steel heddles must run a few weeks to get the best results, but we now expect them to start off like old heddles. In our single thread warp-stop motion, we now use a simple detector mechanism, by which the weaver can easilv find just where the broken thread is, an improvement that will be wel- comed by the weavers. We have called special attention to our Roper head harness motion. We have another form for multiple harness weaving, using a spring top of very simple construction, and some of our customers like it very well. We had some trouble at the start with the frequent breaking of springs, but have hardly any trouble whatever in this line since we have adopted a wooden core bar in the springs. 217 Our double forks started off with records which made us de- cide to recommend them on every loom which we sent out. Cer- tain difficulties became apparent after the looms had run awhile. leading to the invention of a la}- guide, which prevents side mo- tion of the lay after the bearings are worn, and thus preserves the proper co-operation of the forks with the grates. This lav guide is also made so as to hold the lav down in case it should tend to rise from any cause. While this change is very simple, we consider it of great importance. We have also found it pos- sible to improve the double fork mechanism itself, and secure much better results by using a different construction of actu- ator in connection with metal projections attached to the lay, which engage any trailing thread ami prevent the false opera- tion of the fork by a trailing thread. This eliminates thin places, and seems to make the whole construction absolute- ly reliable. Our light J Model loom is undergoing considerable change in construction: for while it is making excellent records, as first sent out, we believe we have noted several chances for improve- ment. We intend to make this the best two-harness loom ever built, and by confining it to two-harness work, we believe we can meet the demand for that class of weaving to better advan- tage than by selling a loom which covers a wider range. In our general loom construction, we have adopted an out- side bearing for the crank shaft, which will make a smoother running loom and practically eliminate crank shaft breakage, a serious trouble with any make of loom. Mr. Roper's let-off has been improved by the application of a hand ratchet mechanism, allowing instant and easy loosening of the warp, with quick increase of tension when desired. Our cut motion has been improved by the additionof special gears which mesh into teeth cut in the cloth roll core bar, so that thev ensure the rotation of this core bar when the cloth is 2l8 first being wound. As the cloth winds, this core bar gradually draws away so that the teeth are not in mesh after the cloth is fairly started. In rocker motions, we have improved the Stearns by downhold lugs, and we have also given careful study to the old '"nigger loot" motion, so called, evolving a pattern which seems to give a most accurate throw for the shuttle. PICKER CHECKS. During the last year or so, there have been quite a few loom picker-check i mentions exploited, some of which were applied to Northrop looms, with the claim that their running was thereby improved. Some of these checks involved the principle of bringing extra pressure on the shuttle in the shuttle box. With our Northrop loom construction, it is necessary for the 2I 9 shuttle to move slightly at times, when the bobbin is trans- ferred. The shuttle, of course, does not reach the same position every time, and it" it goes by the hopper too far, it must slip forward somewhat when the incoming bobbin presses on the shuttle incline. If the shuttle is locked by extra friction at this time, it brings considerable extra strain on the mechanism. The check which we have used for the ten years of our loom introduction has seemed to meet the conditions of usual service, but we have always recognized that it introduced a con- siderable element of expense, the leather wearing continuously. W e therefore spent over six months experimenting with every form of shuttle check and picker-check that our various inven- tors could devise, together with all the checks sent us by outside inventors, or submitted to us for trial. We will not stretch the credulity of our readers by telling them the exact number of de- vices so tested. Suffice to say, that out of the trial, we selected a form which was unquestionably superior to all the rest, so far as the actual results were concerned. When we speak of re- sults, we mean that this check kept the loom running more con- tinuously, and had a better record as to breakage of cops while weaving, cop weaving being peculiarly adapted to test the cushioning effect of check devices. The excessive wear on our early form of check, was due to the harsh blow of the stick in its forward movement. With the present check, no such impact occurs, the return to place being effected by a simple spring (c). The picker when engaged by the shuttle, moves the stick, necessitating the moving of the leather strap through a short space. This motion of the stick acts upon a pivoted lever (a), which by its movement brings pressure against the arm (b) against the strap to still further retard its motion. The resistance is therefore graduated, and the principle seems correct for the results are certainly su- perior. The leather in this check is less expensive, and wears 220 much longer than the -trap in our old check. We have sent this new form out on several thousand looms, and it is giving £ood satisfaction. LOOM TESTS. Outside critic- have sometimes referred to the tests made in our private weaving department as having little or no value for comparative purposes, because we are supposedly using bet- ter yarn and more perfect conditions than are attainable in the average weave room. We wish to correct this impression. by explaining that the warp and filling yarn which we use is not especially prepared for us in any way, but is bought in open market from various mills, and we are confident that if the same yarn were made in a mill which we could control, we should insist on better quality and more uniformity. Our weave room is merely a space partitioned off in our wooden setting-up room, and while we endeavor to secure proper atmospheric con- ditions, v\ e know that we do not secure as good conditions as we should in a weaving room designed for that purpose alone. We see no reason why our records should not be bettered in the average mill where our loom- are run. The one advantage that we d<> obtain i- that of having sufficient extra men so that careful note may be made of every error in operation that does occur. A NEW METHOD OF FILLING HOP- PERS ON NORTHROP LOOMS. In the early days of our loom introduction, we were told by many mill officials, that the records which we were making in our own experimental room would be easily beaten after the mills had taken the looms under their own charge. This we were perfectly willing to concede, and results have proyed that those who use the looms will discover many latent possi- bilities. We found last year that one mill was adopting a method of filling hoppers in which the weavers took a whole handful of bobbins from the filling box atone time, freeing the ends of filling from the bobbins with the other hand, winding off sufficient yarn, and then slipping all the bobbins into the hopper at one operation. This is made easy by the use of our latest hopper construction, haying the new spring discs. Our investigator timed some of these weavers, and found they could put 24 bobbins into a hopper in a minute and a half to two minutes. The speed is facilitated, of course, by having the ends of filling left in proper position when doffing at the spinning frame. We estimate the time required to fill a hopper by the ordinary, single-bobbin method, at about three minutes. The practical advantages of the system were shown by the results being obtained. The weavers were all running 20 looms on ticking, with no help. With common looms on similar goods, they run from 4 to 6 looms in the same mill, paying 32 cents per cut against 11 1-2 cents on the Northrop. The Nor- throp looms were also giving more production and better cloth, with less cost for repairs : in fact, the entire repairs, including shuttles, were not costing over 3 cents per loom per week. The matter struck us as having sufficient importance so that we not only advised all our customers carefully about the advantages of the system, but we have hired expert weavers who understand the system, to teach weavers in various mills how to utilize the new idea. We naturally find more or less opposition from those who are set in their ways, and we also have found that other mills have made improvements of their own over the common system, some of the substitute plans having' considerable merit. A system which accustoms the weaver to placing several bobbins in the hopper at a time has another advantage, in that it uses up less of the weaver's energy in walking back and forth from loom to loom. Weavers are frequently seen putting bob- bins in hoppers when there are only one or two gone. They should economize their effort by learning to wait until a large number of bobbins can be put into the hopper all at once. This latter system will allow them to tend many more looms with no increase of effort. It is the walking around that tires the weaver, and they take a great many unnecessary steps in common practice. ARRANGE3IENT OF LOOMS. Our Mr. C. H. Draper has given considerable study to the problem of arranging Northrop looms so that the weaver shall operate them with the least possible exertion. A great deal of a weaver's time and energv is taken in the moving from one loom to another. The usual weaver plans to get to every loom as it is stopped, so as to start it up at soon as possible, attending 223 to the duties of taking off cloth and putting bobbins in the hop- per while all the looms are running. The problem was figured by taking the average distance a weaver would be obliged to go, by adding the distances from each loom to every other loom, and dividing by the product of the number of looms, multiplied by the number of looms minus one. On the single-alley system, the average distance for 16 print looms figured 19.23 feet. On a double-alley system, 16 looms, the average distance figured 1^.82 feet, a saying of 17.7 per cent. With 24 looms the saying of the double alley system is even greater, the average distance with single-alley being -9.52 and with double alley 21.19, or 2S.2 per cent, saving. With a single-alley system, the average distance between looms increases directly as the number of looms; but with the double-alley system, while the looms are increased 50 per cent., the average distance from loom to loom is increased something like 35 per cent., showing it is easier for a weaver to run a given number of additional looms than it is to run an equal number of original looms. The two-alley system also allows the weaver to move in a circular path while filling flatteries, and thus to be always approaching the batteries which need attention the most: whereas, in the single-alley system, when the weaver reaches the end of an alley, the loom needing attention is at the other end. The problem was also considered in relation to arranging 24 looms in three alleys, eight to the alley. The average distance is practically the same as with two alleys, but there are disadvan- tages in having three separate alleys to move around in. With more than 24 looms there might be advantages in the three-alley system, but so far as our present experience goes, we are not prepared to give a definite recommendation. We do unhesitat- ingly recommend the two-alley system between the limits of 12 to 24 looms to the weaver. 224 SALES. Although we print a complete record of sales to the nearest possible date, a casual reading of the same will hardly give the information which the facts warrant. Sales of improved machinery really prove nothing until the machines themselves have demonstrated their capacity. The real proof of merit is shown when the original trials produce further orders. The greater part of the Northrop looms sold have been on repeat orders, or from parties who had carefully inves- tigated the actual running of the looms in others' mills. We first began to ship looms from our plant in 1S95. It may be interesting to go back and examine the results attained from the very first looms that we sent out. Taking this first year to 1S96, we find that we then sold the Tucapau Mills 320 looms. They have since bought 1439 more, total 1 759 We sold the Queen City Cotton Company 792 looms, and they have since bought 600 more, total x 39 2 Our next order was from the Pacific Mills. 100 looms. They have since bought 20SS more, total 21SS The Merrimack order for 100 looms was entered about the same time. They have recently wanted 204S more for their mills both North and South, total 214S The Amory Mfg. Co. ordered 100 looms. They have since increased, making a total of 6S8 The Lawrence Company took 216 looms. The mill in which they were running was bought entire by the Tremont & Suffolk Company, who afterward bought 1 760 more, total J 9/6 The Grosvenor Dale Company placed an early order for 335 looms. They kept ordering and ordering at vari- ous times; 3282 more in all, total S^ 11 / The Social Company had 196 looms to start with : other orders increase to a total of ^^6 Every Olie of our first eight customers has therefore not only increased their orders, but increased largely. They would hardly continue their patronage had the looms not proved profitable. And we had other customers at this early period, who have since continued their patronage. For instance: First Order. Total Orders. The Pelzer Company 1000 looms. -^\s-2 looms. Gaffney Mfg. Co 1040 •• 1401 Massachusetts Cotton Mills 100 » (both mills) 2833 Lonsdale Company 12 " 2095 Newmarket Mfg. Co 100 " 371 Spartan Mills 1280 " 1880 Dwight Mfg. Co. 16 " 681 We could, of course, add largely to this list, if we referred to more recent examples. We believe those quoted, however, are more pertinent, as it was from the results of our earliest looms that these proofs of satisfaction were derived. We build better looms to-day : their use would give still better satisfac- tion. It may be noted that the mills quoted coyer several states, both North and South. They also cover a wide variety in goods. Their reputation is unquestioned. Their example is certainly worthy of consideration. The Northrop loom has won recognition outside of the United States in spite of the difficulties of foreign introduction. The British Northrop Loom Company Limited has been estab- -6 lished to handle a certain division of Foreign trade. The Soci- ete Alsacienne de Constructions Mechaniques, of Mulhouse, Germany, and Belfort, France, build continuously at both of their establishments. The Ateliers tie Construction Ruti. of Ruti. Switzerland, are manufacturing on various foreign orders for Switzerland. Italy, etc.. and the firm of Isaac Mautner cc Sons of Vienna manufacture for Austria and Hungary. We have sent looms from our own works to Canada. Eng- land. Mexico. Holland. Russia. Japan and elsewhere. LIST OF NORTHROP LOOMS SOLD IN THE UNITED STATES TO JULY i, i 9 o 5 . NAME. PLACE. QJJANTITY. Abbeville Cotton Mills Abbeville. S. C. 940 Acushnet Mill Corp. New Bedford. Mass. 417 Adams Mfg. Co. North Scituate, R. I. 24 Aiken Mfg. Co. Bath. S. C. 3H American Linen Company Fall River. Mass. 100 American Textile Company Cartersville. Ga. 674 American Spinning Company Greenville. S. C. 758 Amorv Mfg. Company Manchester. X. H. 6SS Amoskeag Mfg. Company Manchester. X. H. 1361 Anderson Cotton Mills Anderson. S. C. 724 Androscoggin Mills Lewiston. Maine 204 '■-7 NAME. Appleton Company Aragon Mills Arcadia Mills Asheville Cotton Mills Ashland Company Atlantic Cotton Mills Atlas Linen Company Attawaugan Mills Augusta Factory Aurora Cotton Mills PLACE. HI' A.NTITY. Lowell. Mass. 5*4 Aragon, Ga. IO Spartanburg, S. C. 344 Asheville. X. C. IO Ashland, H.I. 20 Lawrence. Mass. 337 Meredith. X. H. 25 Killingly, Conn. 48 Augusta, Ga. 3- Aurora. 111. 96 L. Bachmann & Company Uxbridge, Mass. Barker Cotton Mills Company Mobile, Ala. Barker Mills Auburn, Maine Bates Mfg. Company Lewiston. Maine Beaumont Mfg. Company ...... Spartanburg, S. C. Belton. S. C. New Bedford. Mass. Berkeley, R. I. Blackstone. Mass. Lowell. Mass. Belton Mills Bennett Spinning Company Berkeley Company Blackstone Mfg. Company Boott Mills Borden Mfg. Co., Richard Fall River, Mass. Botany Worsted Mills Passaic. X. J. Bonnie Mills Fall River. Mass. Bradford Dnrfee Textile School Fall River, Mass. Brandon Mills Bristol Mfg. Corp. Brogon Cotton Mills Brookside Mills Brookford Mills Brower <$: Love Bros. Cabarrus Cotton Mills Greenville. S. C. New Bedford. Mass. Anderson, S. C. Knoxville. Tenn. Brookford. N. C. Indianapolis. Ind. Concord, X. C. 3 2 5 16 1 240 1 256 r 03 2 "3- 652 '4 2000 5 99- 1 366 650 20 5 1- 223 NAME. Cabot Mfg. Company Cannon Mfg. Company Capital City Mills Carolina Mills Centreyille Cotton Mills Chadwick Mfg. Co. Chewalla Cotton Mills Chicopee Mfg. Company Chicora Cotton Mill- China Mfg. Company Chiquola Mfg. Company Clemson College Clifton Mfg. Company Cocheco Mfg. Co. Columbia Mfg. Co. Columbian Mfg. Co. Columbus Mfg. Co. Continental Mills Converse Co.. D. E. Cooleemee Cotton Mills . Cordis Mills Coventry Company Crompton Company PLACE. QUANTITY. Brunswick, Maine 204 Concord. X. C. 426 Columbia. S. C. 216 Greenville, S. C. 160 Centreyille. R. I. ..... 16 Charlotte. N. C. 300 Eufaula, Ala. 40 Chicopee Falls. Mass. 126 Rock Hill. S. C. 1 Suncook. X. II. 89 Honea Path. S. C. 1000 Calhoun Station. S. C. 2 Clifton. S. C. 1000 Dover, X. H. 116 Ramseur, X. C. 69 Greenville, X. H 80 Columbus, Ga. Soo Lewiston, Maine 121 Glendale, S. C. 5^0 Cooleemee. X. C. 1296 Millbury, Mass. 61 Providence, R. I. 2 Crompton. R. I. 2 Dallas Mfg. Company Darlington Mfg. Company Dunbarton Flax Spinning Co. Durham Cotton Mfg. Co Dwight Mfg. Co. Huntsville, Ala. Darlington, S. C. Greenwich, X. V. West Durham. X. C. Chicopee. Mass. s 95 59- 1 300 6S1 Eagle & Phenix Mills Eagle Mills Columbus, Ga. Wi lonsocket, R.I. 328 S Hi) NAME. PLACE. QJJANTITY. Easley Cotton Mills Easley, S. C. iooo Eastman Cotton Mills Eastman, Ga. 150 Edwards Mfg. Company Augusta, Maine 709 Erwin Cotton Mills West Durham. N. C. 1901 Eufaula Cotton Mills Eufaula, Ala. 32 Everett Mills Lawrence, Mass. 660 Exeter Mfg. Company Exeter, N. H. 100 Exposition Cotton Mills Atlanta, Ga. 350 Fairfield Cotton Mills Winnsboro, S. C. 190 Fall River Loom Fixers' Ass'n Fall River, Mass. 2 Falls Company Norwich, Conn. 61 Farnum & Co., John Lancaster, Pa. 12 Farwell Mills Lisbon, Maine 132 Firth. William Boston, Mass. 1 Florence Mills Forest City, X. C. 200 Fulton Bag & Cotton Mills Atlanta. Ga. 1088 Gaffney Mfg. Company Gaffney, S. C. 1401 Gainesville Cotton Mills Gainesville, Ga. ... 1000 Gary & Sons. James S. Baltimore, Md. 3 Georgia School of Technology Atlanta. Ga. 6 Gibson Mfg. Company ... Concord. X. C. 6 Glenn-Lowry Mfg. Co. Whitmire. S. C. 873 (den Raven Cotton Mills Burlington, X. C. 100 Glenwood Cotton Mills Easley, S. C. 96 Gosnold Mills Corp New Bedford. Mass. 800 Granby Cotton Mills Columbia. S. C. 1014 Granitevillc Mfg. Co. Yaucluse, S. C. 362 Graniteville Mfg. Co. Graniteville. S. C. 592 Great Falls Mfg. Co. Somersworth, X. II. 638 Great Falls Mfg. Co.. Rockingham. X. C. 172 NAME. Grendell Mills Grinnell Mfg. Corp. Grosvenor-Dale Company PLACE. QJJANTITY. Greenwood. S. C. 49S New Bedford. Mass. 341 Xo. Grosvenor-Dale, Ct 36 1 7 Hamilton Mfg. Company Hamlet Textile Company Harmony Grove Mills Hartsville Cotton Mills Hathaway Mfg. Company Henderson Cotton Mills Henrietta Mills Hill Mfg. Co. Home Cotton Mills Hope Co.. Phoenix Mill Hoskins Mills Lowell. Mass. 192 Woonsocket. R. I. ^6 Harmony Grove, Ga. 39 6 Hartsville, S. C. 6^0 New Bedford. Mass. 401 Henderson, X. C. 84 Henrietta. X. C. IOI Lewiston. Maine r 4- St. Louis. M( <. 78 Hope. R. I. Soo Charlotte. X. C. sSo Indian Head Mi!!- Alabama Cordova, Ala. Jackson Company Jackson Fibre C . Johnson S: Sons Baltimore, Md. S6 Wamsutta Mills New Bedford, Mass. 300 Richard Borden Mfg. Co Fall River, Mass. 100 Pocasset Mfg. Co. Fall River. Mass. 174 Whitman Mills New Bedford, Mass. [3 Everett Mills Lawrence. Mass. 364 Home Cotton Mills St. Louis, Mo. 102 Pacolet Mfg. Co. Pacolet, S. C. 942 Peppered Mfg. Co. Biddeford. Maine 14 Jackson Co. Nashua. X. II. 60 Swift Mfg. Co. Columbus, Ga. 1 Ninety Six Cotton Mills Ninety Six. S. C. 24 Jackson Cotton Mills Iva. S. C. 350 305 236 LIST OF ATTACHMENTS APPLIED TO OTHER MAKES OF LOOMS TO JULY 1, i 9 o 5 . NAME. Aiken Mfg. Co. Amoskeag Mfg. Co. Androscoggin Mills Appleton Co. Arlington Mills Atlantic Cotton Mills Atlantic Mills Bates Mfg. Co. Boston Mfg. Co. Botany Worsted Mills Cawthon Cotton MillsCo. China Mfg. Co. Continental Mills PLACE Filling- Changer. Warp Mop. Motion. Bath. S. C. I I Manchester. X. H. 10,555 Lewiston, Me 53 53 Lowell. Mass. 47 Lawrence. Mass. 1 13 9 2 Providence, R. I. 1 Lewiston. Maine 26 Waltham, Mass. 300 Passaic. X. L 10 Selma, Ala. 16 16 Sun cook. X. H. 14 <4 Lewiston. Maine Dallas Mfg. Co. Davol Mills Huntsville. Ala. Fall River. Mass. S2 82 Eagle iV. Phenix Mills Everett Mill- Exposition Cotton Mills Fulton Bag & Cotton Mills ( Jibson Mfsr. Co. Columbus. Georgia Lawrence. Mass Atlanta. Ga. Atlanta. Ga. Concord. X. C. 101 773 1 237 NAME. Gosnold Mills Corp. Grinnell Mfg. Corp. Hargraves Mills Hathaway Mfg. Co. 1 PLACE. Filling- Changer. Warp Stop Motion. New Bedford, Mass. -So New Bedford, Mass. 2 Fall River, Mass. 45 2 I New Bedford, Mass. 43 2 Fall River, Mass. 142 6 Clinton, Mass. 2288 King Philip Mills Lancaster Mills Lockwood Company Waterville, Maine Lorraine Mfg. Co. Pawtucket. R. I. Manville Co. Manville, R. I. Manville Co., Globe Mill Woonsocket, R. I. Manville Co., Social Mill Woonsocket, R. I. Mass. Mills in Georgia . Lindale, Georgia Mechanics Mills Fall River, Mass. Merrimack Mfg. Co. Lowell, Mass. Methuen Co. Methnen, Mass. 556 43 409 6 Nashua Mfg. Co. Nashua, N. H. Nanmk'gSteamCottonCo. Salem, Mass. New York Mills New York Mills. X. Y Otis Company Pacific Mills Parker Mills Parkhill Mfg. Co. Pemberton Co. Poe Mfg. Co., F. W. Ware, Mass. Lawrence. Mass. Warren. R. I. Fitchburg, Mas-. Lawrence. Mass. ( rreenville, S. C. 2S 60 '3 ^:v s Warp Fillintf- Stop- NAME. PLACE. Changer. Motion. Rhode Island Co. Spray, N. C. Salt's Textile Mfg. Co. Shetucket Company Stark Mills Stevens Mfg. Co. Susquehanna Silk Mills Samoset Co. .... Tecumseh Mills Fall River, Mass. i i Trainer&SonsMfg.Co..D. Trainer, Pa. i Treinont & Suffolk Mills Lowell, Mass. S 304 Utica Steam & Mohawk Valley Cotton Mills Utica, N. Y. 1 1 Bridgeport, Conn. S Norwich, Conn. 1 Manchester, X. H. 1 Fall River, Mass. 1 1 1 Sunbury, Pa. 7 Valley Falls. R. I. 1 1 Wamsutta Mills New Bedford, Mass. 10 Webster Mfg. Co. Suncook, N. H. i 1 West Bovlston Mfg. Co. Easthampton, Mass. 2 Whittenton Mfg. Co. Taunton, Mass. 4 16 York Mfg. Co Saco, Me 1 69 1,360 17,589 ALSO Complete looms, not on list, shipped to foreign coun- tries or agents, etc. I -7 I 7 Extra Filling-Changers 109 Extra Warp Stop-Motions 85 239 TOTALS. Complete Northrop Looms sold to date, 113,591 Number of Filling-Changers applied,. 113,615 Number of Warp Stop-Motions applied, 129,820 Plain Looms made at or ordered from Hopedale Works, 2,456 The looms changed over include looms made by our licensees in the United States and furnished to mills also in the United States. These figures do not include the many thousand looms made under license in Canada, England, France, Germany. Switzerland, Austria and Hungary. A few of the figures differ from former statements, inas- much as cancellations and exchanges occur more or less. This list is correct at the time of compilation. This volume is intended to contain all the general infor- mation necessary regarding our looms, including all the informa- tion previously published in other catalogues or circulars that is pertinent. We are sometimes asked by overseers or second- hands, to send them books containing numbers and description of our various loom parts in detail. We have such printed lists and are glad to furnish their, to the mills which purchase our looms, but thev are too expensive in character to be generally distributed. Any overseer, or other operative, can probably have access to this list in the mill office, if necessary. ! 4 As soon as this present second edition is exhausted, we shall follow with a third in which the newer devices will be ex- ploited. Any further information regarding looms, or am' of our other products, will be cheerfully furnished on application. To those not fully informed as to the general scope of our busi- ness, we will say that while the Northrop looms are our chief product, we have been introducing cotton machinery improve- ments since 1S16, our line of manufacture before taking up the Northrop loom being devoted to the introduction of High Speed Spindles for spinning frames, Spinning Rings, Spinning Frame Separators, Loom Temples, Warpers, Twisters, Spoolers, Reels, Banding Machines, Balling Machines, etc., etc. We have other literature relating to these products which we will be glad to send on application. DRAPER COMPANY, Hopedale, Mass. July i, igoj. -4 1 TABLE OF CONTENTS. Miscellaneous — Frontispiece, title, etc 1-6 Former literature on the Northrop Loom 7-1 1 The Art of Weaving 12-21 History of the Northrop Loom 22 ~ 2 7 Quotations from advertisements 28-40 The Present .Standing of Our Loom 4 r "45 Hoppers, Thread-cutters, Shuttles, Bobbins, etc. 46-63 Warp Stop-motions ..... 64-69 Devices for Making Perfect Cloth 69-73 Double Fork .....:.... 74~7S Standard Models of Loom Construction 7S~9 2 Loom construction details 93-112 Specification 1 13-1 2 1 Instructions for Running Northrop Loon s 122-173 Cotton Mill Products 174-175 Prices and Profits 176-190 The Labor Question 191-197 Attempts at Competition 19S-201 Patent Infringement and Control 201-206 Addenda 207-223 Sales 224-239 Index 241-244 Memoranda 24S-24S 242 INDEX. Advertisements 28-40 Anti-Bang 90, 96 Art of Weaving ...« 12-21 Bobbins 61-63, 126, 130-134, 171, 193-195 Brake [07-10S, 161-162, 211 Breakage of Filling 125, 129-130, 215 Bunches in Cloth I 3 2_I 33 Bunches on Feeler Bobbins 7 3 ' x 34 Census Reports 174-175, 1S1 Changing Over Looms 92, 193-195 Checks 130, 218-220 Cleaning Looms 45' l( ^6 Cloth Defects 10S, 1 7 1 - 1 73 Cloth Inspection 1 75 Competition 11, 42, 193-195, 19S-200 Construction of Looms 93, 213-214 Cop Looms 54-55i 129-130, 134-135, 173, 215 Cop Skewers 54, 59, 62-63, 135, 215 Cost of Looms 168-169, 1S5 Cost of Weaving 87-100 Cotton Mill Products 174-175 Cut Motion 80, 82, 86, 99-107, 154, 212, 217-218 Dimensions of Looms 1 18-121 Dobby 79, Si, 97 Double Fork 74-75, S9-90, 97, 155-156, 171, 217 Double Pick Cloth 170 Drawing-in Frame 112, 164 Drop Wires 67-68, 141-144, 173 Feeler (or Mispick Preventer) 62, 69-73, 85, 149-15 1, 172, 200 Feeler Thread-Cutter 72-73, 150-151 2 43 Filling Fork 155-156 Floats 45, 66, 171 Foreign Loom introduction 201, 225-226 Hand Loom 12-17 Harness Cams 146-147 Heddles 64-67, 135-136, 13S-I-IO, 145, 216 History 1 1-43 Hopper 46-49,52-57, 76-89, 122-124, '7 1 ! 221 ' 222 Instructions for Running Northrop Looms 122-173, 221-223 Knots in Warp 146 Labor Question 191 -197 Labor on Plain Loom 43~44> r 93 -I 95 Lay I56-157, J 59' l6 5i 2I 7 Lay Adjustments 123-124, 150, 157, 159, 217 Let-off 77, 94-96, 151-152, 172, 212, 217 List of Inventors 2, 203-206 Literature on Northrop Loom 7" 11 ' 26-27 Litigation 1 1, 199-200 Long Bobbin Experiments 193-195 Loom Adjustments 163 Loom Arrangement 222-223 Loom Equipment 169-170 Loom Power 165 Loom Seats in, 20S Misthread Stop-motions 50-^1 , 1 28 Misthreading 50, 60, 127-1 29 Models of Looms 7^"9 2 < 211, 217 Number of Looms per Weaver 77, 80, S5. 87, 94, 97, [83, 18S-1S9, 193-194. 197 Patents 2, 17-19, 63, 177, 199-206 Patent Control 11, 200-206 Patent Infringement 199-200 Percentage of production, So, 167. 175, 178-180, 1S3. 187-189, 197 2 44 Plain Power Loom. 14-21,43-44, 6$. 90-92, 182, 191-192. 193-19^ Plan of Works ... 6 Press Notices 7-10 Prices 176, 185, 190 Print Cloth 15-16, 74, 82, 89, 97, 1S7 Product per Operative 14-16, 25. 117. 167. 177, 197 Profits by use of Northrop Loom 10c, 176-190 Protector 126, 161 Ree 21-25, 19S-200 Shuttle Guard 1 10 Shuttle Position Detector, 50-51. 56-57, 72-73,86, 122-123. I2 $. 216 Sizing Warp . 164-165 Slack threads * 66, 144 Specifications 11 3-1 21 Speed 14-16.90, 165. 16S, 175, iSS. 206 Take-up 77-S0. S2. 9S-107. 153-155. 172 Temple Thread-cutter ^o-^i. 14S-149. 216 Thin-place Preventer 74, 109, 217 Transferrer 49. 53, 57, 123 Warp Beams 1^2 Warp Breakage 66-67, I2 5< x 4 ' J 45- x 93« 2I 6 Warp Stop-motion 25, 4^-45. 64-69. So. S6, 112. 135-144. 193-195, 200. 20S-209. 212, 216 Waste 72, 169 2 45 .MEMORANDA. 246 MEMORANDA. 247 MEMORANDA. -4 s MEM OR AX DA.