DR. SAMUEL C. BUSEY 
 
 COLLECTION. 
 
^f 
 
 
 ♦♦ 
 
 o 
 
 o 
 
 o • 
 
 
 o o 
 
 g 
 
 o Cl f^ 
 
 o o 
 
 o " ; t; 
 
 «5 C> •! 
 
 ei « c .H 
 
 « JL,' tH t-. I4 
 
 K ♦* r-J f-' f? 
 
 ^ ^ S *^ 
 
 f ■ t- 
 
 ^ ,^ o « r,; I., 
 . f3 *> J, I c: 
 
 H • S ^ r-( 1^ 
 
THE LIBRARY 
 
 OF 
 
 THE UNIVERSITY 
 
 OF CALIFORNIA 
 
 PRESENTED BY 
 
 PROF. CHARLES A. KOFOID AND 
 
 MRS. PRUDENCE W. KOFOID 
 
. )■ 
 
 
 
 PHYSIOLOGY AND PATHOLOGY 
 ~ OF THE BREAST. 
 
CONTRIBUTIONS TO THE 
 
 PHYSIOLOGY AND PATHOLOGY 
 
 OF THE 
 
 BREAST 
 
 AND ITS LYMPHATIC GLANDS. 
 
 BY 
 
 CHARLES CREIGHTON, M.B. 
 
 DEMONSTRATOR OF ANATOMY IN THE UNIVERSITY OF CAMBRIDGE. 
 
 / 
 
 ^Jt . c<k ♦ 
 
 WITH ILLUSTRATIONS. 
 
 Hontion : 
 
 MACMILLAN AND CO. 
 
 1878 
 
 [All Rights reserz'ed.] 
 
Cambrrtge : 
 
 PRINTED BY C. J. CLAY, M.A. 
 AT THE UNIVERSITY PRESS. 
 

 O 
 
 O 
 
 PDBNC LIBBABY, 
 
 la/TTf-r OR AlWIV 
 
 DEC 27 1909 
 ton, D. 0, 
 
 The inve^^^ttions describe^^^his volume were 
 undertaken fnr fm^^WrrifrTr'^PfpTrtiiiriif of the Privy 
 Council by way of inquiring into the causation of malig- 
 nant tumours. An earlier investigation, on the origin of 
 secondary tumours in the liver, formed the subject of a 
 paper in the Medical Officer's Keport for 1874. This paper 
 does not fall within the plan of the present work, and I 
 have not reprinted it. The five chapters on physiological 
 and pathological processes of the breast are reprinted, 
 with additions and corrections, from the Report of 1875, 
 and the chapter on Tumour-Infection of Lymphatic Glands 
 from the Report of 1876. The chapters on development 
 were originally published in the Journal of Anatomy and 
 Fhysiology, Vol. xi., 1876. 
 
 The investigations were made during Mr Simon's term 
 of office, and I enjoyed from time to time the opportunity 
 of conversing with him on the subjects of them. The 
 scope of pathological inquiry which he set before me was 
 of the widest kind, and he strongly encouraged the physio- 
 logical direction that the work has taken. I desire grate- 
 fully to acknowledge his never failing interest in the 
 details of my investigation during its progress, and the 
 assistance that I have derived from his criticisms. 
 
 Cambridge, 
 
 28th Feb. 1878. 
 
 c. 0. b 
 
 M371710 
 
CONTENTS. 
 
 PAGE 
 
 Introduction 1 
 
 PART I.— PHYSIOLOGICAL. 
 CHAPTER L 
 
 THE PERIODICAL INVOLUTION OP THE BREAST. 
 
 Structure of the Breast in the Lactation Period— luvoluted or Resting 
 State — Intermediate Stages of the Upfolding Process — The Secretory 
 Changes in the Epithehum, as exhibited during tlie Subsidence of 
 the Function — The Secretion, a Collateral Product of Endogenous 
 Cell-Formation— Cellular Waste Products of the Secretion — Pig- 
 mented Waste Cells — Successive Renewals of the Epithelium through- 
 out the Subsidence of the Fmiction, as well as in the Lactation 
 Period — All the Waste Cells theoretically considered to be Products 
 of Endogenous Cell- Formation — Cellular Waste Products miavoid- 
 able in a Subsiding Secretion 9 
 
 CHAPTER II. 
 
 THE PERIODICAL EVOLUTION OF THE BREAST. 
 
 Relation of the Periodical Unfolding to Pregnancy — Material of the In- 
 vestigation — First Stage of Evolution — Second Stage of Evolution — 
 Third Stage of Evolution — Unfolding of the Structure correlated 
 with Increasing Intensity of the Fimction ; Dualism of Structure 
 and Function — -Various kinds of Waste Products corresponding to 
 Stages of Evolution — Parallelism with the Process of Involution . 38 
 
 CHAPTER IIL 
 
 THE LYMPHATIC GLANDS OF THE BREAST IN CONNEXION WITH THE DISPOSAL 
 OF ITS CELLULAR WASTE. 
 
 Summary of the Waste Products of the Waxing and Waning Secretion- 
 Lymphatic Glands of the Breast as Receptacles for the Waste Cells 
 — Summary of Opinions on the Structure and Function of Lym- 
 
viii CONTENTS. 
 
 PAOE 
 
 phatic Glands — Pigmented Waste Cells in the Lymph Sinuses — Re- 
 duction of Waste C.^ells by the Reticulum of the Lymph-Siniises — 
 Other modes of their Reduction — The Denuded Waste Cells enter 
 the Lymphatic Follicles as Ljinphoid Cells— The Follicular Cells are 
 contained in the Adventitia of Veins — Utilisation of Cellular Waste 
 — Leucocytosis of Pregnancy -"J? 
 
 CHAPTER IV. 
 
 THE DEVELOPMENT OF THE BREAST. 
 
 The Received Theory of the Development of the Breast — Theory of the 
 Development of Glands by Recession or Inversion of the Surface 
 Epithelium — Goodsir's Views on the Development of Glands— The 
 Mamma of the Newborn Guinea-pig — Resemblance between it and a 
 Cluster of Fat-Lobules — Inguinal Fat-Bodies in the Kitten — The 
 Stages of their Development — Course of Development of Mammary 
 Acini in the Guinea-pig — Parallelism between the Development of 
 Fat-Lobules and Mammary Acini — Comparative Anatomy of the 
 Breast ; the Mammary Glands of the Monotremata — Nipples and 
 Branching Ducts are the latest Acquisitions in the Succession of 
 Animals, but the first Structures to appear in the Embryonic De- 
 velopment of a Higher Mammal — Reversal of the Order of Acquisition 
 explained according to Mr Spencer's Hypothesis of Direct and In- 
 direct Development — The Development of the Ducts— Conclusions 
 of the Inquiry 83 
 
 CHAPTER V. 
 
 THE DEVELOPMENT OF THE MAMMARY FUNCTION, 
 
 The Development of Function, a New Subject — Imperfect or Rudimentary 
 Mammary Function in the Monotremata — Existence of the ^Mamma 
 in the Male Monotremes (note) — The ^lammae of the Monotremes 
 resemble Fat-Bodies in their Expanded and in their Contracted 
 State — The Periodicity of the Mammaiy Fimction, its Earliest 
 Characteristic — The Development of Function in the Individual 
 Embryo — The Milk of the Newborn — The Periodical Processes of 
 the Mature Breast are Repetitions of its Development — Dualism of 
 Structure and Function in Developmental Processes . . .109 
 
CONTENTS. 
 
 PART II.— PATHOLOGICAL. 
 
 CHAPTER VI. 
 
 PATHOLOGICAL TROCESSES OF THE BREAST. 
 
 PAGE 
 
 Summary of the Normal Cellular Processes — Endogenous Cell-Formation 
 iu Pathology — ]\Iaterial of the Pathological Investigation ; Tumours 
 in the Bitch — The First Step in Tumour Formation is a Spurious 
 Evolution from the Eesting State — Accumulations of Pigmented 
 Waste Cells in the Secreting Structure^Collections of the same in 
 the Stroma of the Breast — Criticism of the Connective-Tissue 
 Theory of Skirrhous Tumours — Tumours formed by Waste Cells of 
 the Nuclear kind — Tumour- Formation in the Breast is primarily a 
 Disorder of Function ; Dualism of Structure and Function in Morbid 
 Processes 121 
 
 CHAPTER VII. 
 
 PATHOLOGICAL PROCESSES OF THE BREAST — Continued. 
 
 Morbid Processes corresponding to the Mucus-producing Stage of the 
 Rising Function — Intra-acinous Myxomatous Formations — Descrip- 
 tion of a Particular Case — Oblong or Spindle-shaped Cells as Epithe- 
 livmi — The Same outside the Secreting Structure as Waste Cells — 
 Myxo-sarcomatous Tumour of the Breast originating in their Extra- 
 acinous Accumulation — Criticism of the Connective-Tissue Doctrine 
 — Cartilaginous Transformation of IManimary Tumours in the Bitch 
 — Evidence that the Epithelial Cells and the Glandular Force are 
 concerned — Papillary and Polypous Growths in the Acini and Ducts 
 —Relation of Epithelial Cells to Connective-Tissue forms of Tumour, 
 not one of Equivalency but of Metaplasia — Description of Eight 
 Cases of Mammary Timiours in Women — Pre-climacteric and Post- 
 climacteric Tumours of the Breast — The Effacement of the Breast at 
 the Climacteric Time of Life — General Conditions of Tumour-For- 
 mation in the Breast — Contrast between Disturbances from the 
 Resting State, and Interferences with the Perfect Lactation ; In- 
 flammation and Abscess — Contrast between Mammary Tumours 
 within the period of Sexual Vigour, and after the time of Sexual 
 Effacement — Summary of the CJeneral Liability of the Breast to 
 Tumour-Disease 149 
 
X CONTEXTS. 
 
 CHAPTER VIII. 
 
 TUMOUR-INFECTION OF LYMPHATIC GLANDS. 
 
 PAGE 
 
 Primary Tumours viewed either as Products of Functional Disturbance, or 
 as Individual Things ; the Individuality of a Tumour evidenced by its 
 Infectiveness — The Material of the Investigation — Writings on Lym- 
 phatic-Gland Infection — Tumour-Infection in the Lymphatic Glands 
 of the Breast — The Advantage of beginning with Mammary Cases — 
 Description of a Case in the Bitch — Mimicry of the Primary Struc- 
 ture produced in the Lymphatic Gland by carving the Pattern out 
 of the Lymphatic Parenchyma — Varieties of the Process of Adapta- 
 tion — The Points of Failure of the Normal Lymphatic Function — 
 Transport of Primary Tumour-Cells to the Lymphatic Gland — Difter- 
 ences in the Process of Infection depend upon DiflFerences in its 
 Intensity — Endogenous Cell-Formation in the Infective Tumoiu" 
 Process — The Endogenous Change ; in the Cells generally, or in 
 certain Representative Cells — Comparison of the Infectiveness of 
 Tumours to the Spermatic Influence — Contrast between the mode 
 of Origin of Primary and of Secondary Tumours — The Begetting 
 Influence of the Primary Tumour-Disease is not always one and 
 indivisible 178 
 
LIST OF WOODCUTS. 
 
 FIfJ, PAGE 
 
 1. Fully expanded mammary acinus . 10 
 
 2. Varieties of polyliedric epithelium 11 
 
 3. Involuted or resting state of mammary lobules ..... 13 
 
 4. Nuclear condition of the epithelium in the resting state . . . .14 
 a. Lobule of the mamma of a cat nine days after the end of lactation . 15 
 
 6. The peculiar forms of epithehum in the middle period of involution . IG 
 
 7. Acini from the udder of a ewe shortly after the end of lactation . . 17 
 
 8. Large granular and vacuolated cells in the mammary acini shortly after 
 
 the end of lactation 18 
 
 9. Portion of mammary lobule and interlobular tissue, showing the nuclear 
 
 waste cells of the subsiding secretion . 26 
 
 10. Lobule with pigmented waste cells in the acini and in the fibrillar tissue. 29 
 
 11. Lobule with pigmented cells in the form of columnar epithelium . . 30 
 
 12. Two acini in an early stage of periodical evolution 41 
 
 13. Pigmented waste cells passing into a duct 42 
 
 14. Pigmented waste cells collected in the fat-tissue surrounding the mamma. 43 
 
 15. Middle stage of periodical evolution 45 
 
 16. Condition of the interlobular tissue in the middle period of evolution . 47 
 
 17. Late, or mucus-producing stage of evolution 49 
 
 18. Lymphoid cells collected in the wall of a vein— rudimentary follicle of 
 
 lymphatic gland . 73 
 
 19. Part of a mammaiy IjTnphatic gland, the lymph-sinuses containing 
 
 large granular cells from the breast, and the trabecule crowded with 
 
 oblong cells 74 
 
 20. From a mammary lymphatic gland, showing that the lymphoid cells are 
 
 collected in the adventitia of the smaller veins 78 
 
 21. Mammary lobule in a state of morbid evolution 128 
 
 22. Pigmented cells of the mamma of abnormal shape 129 
 
 23. Pigmented waste cells in the form of villus-like outgrowths of the mucous 
 
 membrane 130 
 
 24. The same accumulated in papillary processes 131 
 
 25. Pigmented cells infiltrated into the fibrillar tissue (from a tumour in the 
 
 bitch) 134 
 
 26. The cells in the stroma, from a case of skirrhus of the breast in a woman 135 
 
xii LIST OF WOODCUTS. 
 
 no, PAGE 
 
 27. Nuclear condition of tlic mammary opithelium in tlif midille period of 
 
 normal evolution 140 
 
 28. Villus-like outgrowths of nuclear cells within the acini .... 142 
 2t). Nuclear cells forming several layers on the wall of an acinus . . . 143 
 80. Trabecular columns of nuclear cells occupying the acinoiis space . . 144 
 
 31. Similar arrangement of polyhedric nucleated epithelium from a cancer 
 
 of the male breast .......... 145 
 
 32. Mammary acini lined by columnar miicus-yielding epithehum . . 151 
 
 33. Mammary acini lined by several layers of columnar and spindle-shaped 
 
 epithelium . , 155 
 
 34. Mammary tumom- made up of extra-aciuous collections of oblong and 
 
 spindle-shaped cells 158 
 
 35. Group of acini illustrating one variety of the cartilaginous transformation 161 
 
 36. Group of acini with their cells vacuolated; the vacuoles filled by a 
 
 firm and hyaline substance 1C3 
 
INTRODUCTION. 
 
 The breast is one of a group of organs in tlie body, belonging 
 for the most part to the reproductive system, which have not as 
 yet come to any considerable extent within the scope of the 
 experimental physiology. The breast may be considered to be a 
 sufficiently typical organ of secretion, but the questions relating 
 to the nature of secretion have been tried almost exclusively 
 on other secreting glands, notably on the salivary glands, which 
 come easily within the reach of experiment. But although the 
 breast has been almost neglected by experimental physiologists, 
 and may indeed be viewed as a secreting organ ill-adapted for 
 the application of their method or for the illustration of their pro- 
 blems, yet the circumstances of the breast present an unrivalled 
 opportunity for the study of the process of secretion. The end 
 of experiment may be said to be to vary the circumstances. From 
 the variations so produced, the facts are gained on which to make 
 an induction. As regards the breast, it stands alone among the 
 secreting organs of man's body in presenting to the observer 
 a series of variations which may be called natural or spontaneous 
 in contrast to those that are produced in other glands artificially 
 by experiment. In so far as the circumstances of the breast vary 
 of themselves, under conditions that are well known, the effects 
 upon the secretion have a regularity and certainty which is often 
 wanting to the effects of experiment. Apart from their uniformity, 
 the variations of the mammary secretion are in themselves of the 
 most instructive kind. It is perhaps not altogether superfluous 
 to point out that experiment is nothing more than a means to 
 c. c. 1 
 
 % 
 
2 INTRODUCTION. 
 
 an end, and that the scope of physiology includes all generali- 
 sations as to function, whether they have been arrived at by 
 experiment or without experiment \ 
 
 The instructive variations in the secretory process of the 
 breast depend on the circumstance that the function is periodically 
 called into play, and periodically subsides. Lactation, properly 
 so called, begins at the birth of the offspring, and, among the 
 various mammals, lasts for a certain number of weeks or months. 
 But the function of the breast is not suddenly called into play at 
 the time of parturition, nor does it suddenly cease at the end of 
 suckling. The function subsides gradually, and when the next 
 pregnancy is established, it returns by still more gradual stages to 
 the full intensity that it possesses at or just after the birth. It is 
 in the stages of gradual subsidence, of still more gradual reawaken- 
 ing, and in the remarkable parallelism between them that one finds 
 the valuable means of generalising upon the mammary secretion. 
 Hand in hand with the subsidence of the function, there goes on 
 an upfolding of the structure, the result of which is visible 
 externally in the diminished size of the organ. The revival of the 
 functional activity is in like manner accompanied by an unfolding 
 of the structure, the outward sign of which is the increasing size 
 and turgescence of the organ from the early months of pregnancy to 
 the time of delivery. The variations in the intensity of the secre- 
 tory force as measured by its products correspond to changing as- 
 pects of the secreting acini. In following the processes of upfolding 
 and unfolding structure, of subsiding and reviving intensity, one 
 is sometimes apt to think that the mysterious dualism of structure 
 and function comes near to being surprised of its secret. 
 
 Notwithstanding the attractive nature of the problems here 
 offered for study, the periodical processes of the breast have been 
 hitherto almost entirely undescribed. Langer has indeed referred 
 to them, and in his memoir^ on the breast, published in 1852, he 
 expresses the intention of afterwards entering more fully into the 
 subject. But in an article' on the breast, published in 1870, he 
 
 ^ " Observation is finding a fact, Experiment is Tnaling one. The worth of the 
 fact depends on what it is in itself, and not on the manner of obtahiiug it. " 
 
 Bain's Logic. 
 ^ Denkischriften dcr Wiener Akademie, 1852. 
 3 Strieker's Handbuch. 
 
INTRODUCTION. 3 
 
 goes no farther than to refer to the upfolding process under the 
 name of the involution of the gland, and to describe briefly and in 
 general terms the resting state of the gland, or the condition in 
 which it remained at the end of involution. I do not find that 
 these questions relating to the breast have been either fully stated 
 or at all worked out by any other author. 
 
 The term "involution," used by Langer for the upfolding pro- 
 cess, appears to be very suitable, and will be used in the sequel, 
 with the addition of "periodical" to distinguish it from the final 
 atrophy or disappearance of the secreting structure that occurs in 
 women at the climacteric years, as part of the general " sexual 
 involution." If the word " involution" be used for the upfolding 
 process, the correlative term for the unfolding process cannot well 
 be anything but " evolution." Both terms will at least have the 
 advantage of being applied in their literal sense, and the evolution 
 being further defined as " periodical " may perchance escape the 
 danger of being mistaken for any of the other and sometimes 
 lax applications of that word\ 
 
 The first two chapters will contain an account of those two 
 processes of the breast. The third chapter treats of the lymph- 
 atic glands near the breast as appendages of the secreting gland. 
 While the chapter on lymphatic glands is an integral part of the 
 investigation on the mammary secretion, it is at the same time 
 an independent contribution to the knowledge of the structure and 
 function of lymphatic glands ; and it will be found to contain a 
 comparatively new exposition of their function. 
 
 The fourth and fifth chapters contain a new account of the 
 development of the breast, and some considerations on the develop- 
 ment of its function. The development of the function of an 
 organ is naturally bound up with the development of its structure; 
 it is the same dualism of structure and function that has been 
 referred to. But the function in its development is as capable of 
 independent treatment as the function in its maturity; and I 
 
 1 Astley Cooper {Anatomy of the Breast, London, 1840) applies the term evolution 
 to a group of developmental processes, including the iirst formation of the gland 
 in the embryo, the changes that it undergoes in both sexes up to puberty, and the 
 completion of its development or expansion in the primipara. He contrasts also 
 the full state of lactation with other states of the gland. (See a reference to one of 
 his observations in the second chapter.) 
 
 1—2 
 
4 IXTRODUCTIOK. 
 
 shall bust justify my introductiou oi' it as a separate topic by 
 quoting the conclusion that I have arrived at. The embryonic 
 cells, in order to become the secreting cells of the mammary acini, 
 go through a cycle of changes ; and the changes that they undergo 
 are precisely those that the cells of the mature organ undergo in 
 producing their periodical secretion. The function of the breast 
 is therefore said to be a sustained repetition of its development. 
 The first "period" of the breast is its development. 
 
 Chapters VI. and VII. contain the application to disease of 
 the physiological principles established in Chapters I. and II. 
 They treat of morbid physiological processes of the breast, and they 
 endeavour to show from the examination of a series of cases that 
 the common and malignant tumours of the breast, as well as some 
 of the rarer forms (Chapter VII.), owe their origin to functional 
 irregularities of the gland. The varieties of cancers of the breast, 
 hard and soft and the like, have necessarily been kept in view ; 
 but the tumours are here discussed, not with reference to their 
 subdivision on points of structure, but always with a view to 
 making out the particular departure from the normal process of 
 the gland to which their existence in each case is due. In other 
 words, it is the excitation {Reiz) that is chiefly considered. In a 
 former paper* I defended the use of the word "cancer" as a 
 general name for all malignant tumours. It is here used in that 
 sense ; and it can be so applied with strict propriety to a class of 
 tumours described in the sequel, which from the shape of their 
 cells might be called sarcomata, but which are none the less the 
 product of functional disturbance of the secreting parenchyma. 
 
 The last chapter is an application to pathology of the facts 
 arrived at in Chapter III., (m the function of lymphatic glands. 
 The lymphatic glands of the breast being viewed as functional 
 appendages of the secreting gland, it is possible to treat their 
 invasion or infection in cases of tumour-disease of that or^an as 
 being in a very particular way incidental to the health-relation- 
 ship subsisting between them. This physiological clue to the 
 infection of mammary lymphatic glands has been made use of, in 
 the latter part of the chapter, to illustrate by resemblance and 
 by contrast the general question of tumour-infection both as re- 
 
 ^ " On the Histogenesis of Secondary Tumours in the Liver,'' in the Report 
 of the Mfdical Officer of the Privy Council for 1874, pp. 95, 9C. 
 
INriiODUCTIOX. 
 
 gards the primary and secondary tumours of other epithelial 
 organs, and also as regards the primary and secondary tumours ot 
 connective- tissue parts. 
 
 At the beginning of this preface I contended that it was 
 possible to study the functional processes of a gland so far without 
 the modern aid of experiment. I might now in like manner point 
 to the pathological conclusions of my investigation as showing the 
 utility of a kind of physiology that is neither physiological physics 
 nor physiological chemistry. The cellular pathology is nowhere 
 more triumphant than in the region of tumours, and I believe 
 that for every organ or part of the body that is subject to the 
 tumour-disease — not to speak of other diseased processes — there 
 exists a cellular physiology as distinguished from histology, which 
 will always demand a fair share of attention at the hands of the 
 medical profession. However convenient and however philo- 
 sophical it may be in modern physiology to ignore the indi- 
 vidual elements of the body, and to view them in the abstracted 
 form of a protoplasm upon which forces play or in which chemical 
 changes take place ; yet, when we come to the everyday pro- 
 cesses of disease, the elements of the body are constantly asserting 
 their individuality, and in the case of tumours they assert it in 
 the most alarming way. It appears to me, therefore, that phy- 
 siologists owe greater consideration to the cellular unit of the 
 body, and to the spontaneity and self-governing properties of 
 the cell. 
 
PART 
 
 PHYSIOLOGICAL. 
 
CHAPTER I. 
 
 THE PERIODICAL INVOLUTION OF THE BREAST. 
 
 It will be convenient to give, at the outset, some account of 
 the mamma in its state of perfect expansion and full secretory 
 activity, not with a view to bringing forward new points, but for 
 the sake of illustrating the successive changes that the organ under- 
 goes. The characteristic appearances of a gland well advanced in 
 involution will next be described. This condition of the orsfan 
 has been found, in the dog, cat, sheep and rabbit, to fall between 
 the fourth and sixth weeks after lactation has ceased, and it may 
 be taken to be the resting state of the mamma. The entire series 
 of intermediate changes from the end of suckling to the resting 
 state will then be entered on, and this part of the investigation 
 will afford an opportunity for discussing what appears to be a law 
 of the secretion. 
 
 During the period of lactation, and for a few days both before 
 and after it, the mamma is found to be a finely lobulated organ, of 
 a pinkish colour, succulent, of uniform consistence, and having 
 generally a greater resemblance to the pancreas than to the 
 salivary glands. 
 
 In the dog, cat, and rabbit, the chain of glands extends along- 
 each side from the groin to the middle of the thorax ; except in 
 one case in the cat (a primipara) the chain of glands has been 
 found to be confluent, forming a single long strip on each side. In 
 the case mentioned as an exception there were found, correspond- 
 ing to the five teats, five perfectly distinct circular glands with 
 bands of connective tissue stretching between them. In the 
 
10 
 
 THE PERIODICAL INVOLUTION 
 
 [Pt. I. 
 
 general case where the glands are confluent, the same appearance 
 of isolated circular areas is produced by the several sets of ducts 
 (where these are visible) radiating towards each teat. 
 
 In the dog, cat, and rabbit there are from four to five teats on 
 each side, in the sheep there are two pairs, a large and a small, 
 and in the guinea-pig there is a single pair. In the animals that 
 have the extended chain of glands, the anterior two-thirds of the 
 chain is flat and thin, being about two inches in width, and a line 
 or more in thickness ; towards the inguinal region the gland be- 
 comes much narrower and thicker, and tapers to a rounded end at 
 the symphysis pubis. The upper surface of the gland is closely 
 adherent to the skin throughout ; the lower surface is separated 
 from the abdominal wall by a thin layer of loose connective tissue 
 in the anterior part of the chain, and in the inguinal region by a 
 cushion of fat of variable thickness. Small bundles of striped 
 muscular fibre are met with at various points in the gland, running 
 in a longitudinal direction. Their occurrence is, however, not 
 uniform, and there is no regularity in their distribution; they 
 have the appearance of being detached bundles of the subjacent 
 muscle which have, during development, become included in the 
 gland substance. 
 
 In the minute structure of the fully expanded gland, the fol- 
 lowing points are to be noted. Each lobulus is composed of a 
 large number of closely packed acini, for the most part of round 
 shape and uniform size, while those at the borders or corners of 
 
 the lobule may be larger and elon- 
 gated. The lobules are separated from 
 one another by very thin tracts of loose 
 connective tissue ; the acini making up 
 a lobulus are bounded by the encircling- 
 blood-vessels and their adventitia. The 
 grouping of the acini round the ducts 
 cannot well be made out in sections 
 of the fully expanded gland. The floor 
 of an acinus in section is covered by 
 a mosaic of polyhedric epithelial cells, 
 usually to the number of 15 or 20, 
 while in the larger elongated acini as 
 many as .30 may be counted. The cells are usually pentagonal or 
 
 Fig. 1. 
 
 Fully-expanded acinus of 
 
 the uiaiimia, showing mosaic 
 or (lol.vlirili-ic epithelium. Mag- 
 iiilit il .JUO diameters. 
 
Ch. I.] OF TUB BREAST. \\ 
 
 hexagonal, and the corners are sometimes rounded. In each cell 
 there is a central round nucleus which colours brightly with 
 the staining fluid (purple with logwood), and a broad fringe 
 of protoplasm which stains less deeply. The nucleus is on an 
 average not more than one-third of the entire breadth of the cell. 
 Fig. 2 shows variations in the size of the nucleus relative to the 
 surrounding protoplasm. In a 
 profile view of an acinus the 
 epithelium appears as a circlet 
 of oblong cells in which the 
 nucleus at the centre occupies 
 almost the entire thickness of 
 the cell. The mammary epi- 
 thelial cell may therefore be ,, 
 
 '' Mammary epithelium, showing variations 
 
 described as a flattened poly- in the relative amount of nucleus and cell- 
 
 ^ -^ substance. Magnified 300 diameters. 
 
 hedric body, with a thickness 
 
 about one-half of its breadth. The substance of the nucleus is 
 apparently homogeneous, with a deeper line of staining round 
 the margin ; a nucleolus is not always prominently seen. The 
 broad fringe of protoplasm is more or less granular or hyaline 
 varying according to conditions that do not require to be discussed 
 at the present stage. By special methods of preparation the 
 meshwork of the basement membrane may be demonstrated. It 
 is seen to be a regular arrangement of delicate fibres, forming 
 polyhedric spaces in which the epithelial cells are set. It may be 
 added that the best preparations of the perfect mammary epi- 
 thelium are obtained, not when the animal is in milk, but two or 
 three days before parturition. 
 
 Turning now to the description of a gland well advanced in 
 involution, the extent of the changes that constitute the involu- 
 tion process will become at once apparent. The particular case of 
 a bitch killed five weeks after lactation had ceased may be taken 
 as an example. The entire chain of glands is considerably re- 
 duced in thickness, and the anterior end of the chain so much so 
 as to be quite membranous. In the thinnest part of the chain, 
 the ducts, which are not to be seen in the state of full expansion, 
 are now visible as large round cord-like structures radiating from 
 the periphery to the nipple. The inguinal portion of the chain 
 now rests on a thick cushion of fat ; in certain cases this cushion 
 
12 THE PERIODICAL INVOLUTION [I't. 1. 
 
 has been found to be as much as three-quarters of an inch in 
 thickness, while the gland-tissue formed, as it were, a mere covering 
 drawn over it. The cushion of fat is also observed to contain 
 small lymphatic glands embedded in it. On cutting into the 
 gland thus reduced in thickness, its substance is found to be 
 tough, and to consist of large tracts of not very dense fibrous 
 tissue, interposed between small round or linear areas of the in- 
 voluted glandular tissue. The latter have the usual colour and 
 the soft or parenchymatous section of glandular substance, and in 
 the plane of section they incline to stand out slightly above the 
 level of the fibrous stroma. In the greatly flattened or spread - 
 out kinds of glands, as in the dog, cat, and rabbit, the longi- 
 tudinal section has a stratified appearance, from the arrangement 
 of several involuted lobuli in a long narrow strip, with correspond- 
 ing parallel tracts of connective tissue. In the more compact 
 glands of the sheep and guinea-pig, the islands of glandular tissue 
 are more uniformly scattered over the plane of section. It some- 
 times happens that the tracts of connective tissue between the 
 lobules are occupied by fat. Where this is the case, the contrast 
 between the true glandular and the other tissue is very marked, 
 and the stratified or mottled appearance more striking. The 
 glandular tissue might be estimated roughly to make up less than 
 one-half of the entire bulk of the organ. The section of the in- 
 voluted gland shows also numerous large ducts and blood-vessels 
 cut across. 
 
 On microscopic examination the ducts and blood-vessels are 
 always conspicuous. In sections of the fully-expanded glands 
 they are seldom encountered ; but the great retraction of the 
 glandular substance in the involuted state enables them to become 
 prominent. The fibrous tissue, as has been stated, is sometimes 
 occupied by fat. In most cases, however, the tissue between the 
 lobules is purely fibrillar or coarse fibrous, and the greater or less 
 amount of fat seems to follow the individual peculiarities of the 
 animal. The most striking instance of fat formation occurred in 
 a rabbit. A portion of the mamma, four days after the young 
 were taken away, was found to have the uniform parenchymatous 
 appearance of the fully expanded gland. When the animal was 
 killed ten days later, the glandular tissue Avas found to be greatly 
 retracted or involuted, and the involuted lobuli or groups of lobuli 
 
Cm. T.] 
 
 OF THE BREAST. 
 
 13 
 
 to be separated by broad tracts of pure fat-tissue. This rapid 
 development of fat in the rabbit is in accordance with the obser- 
 vations of Toldt' and of Czajewicz^ as regards the rapid disappear- 
 ance or new formation of the fat-tissue generally in that animal. 
 
 The attention is next directed to the minute structure of the 
 gland-tissue proper in the state of involution. The small areas of 
 glandular substance are found grouped with some regularity 
 round the larger ducts. Fig. 3 shows a large central duct, send- 
 
 FiG. 3. 
 
 
 
 ^.'-.>ic:^-^ ^-f^r^v.^^a. ?M'°MifSy " 
 
 From the mamma of a bitch five weeks after the end of lactation. 
 Characteristic appearance of involuted lobules grouped about a duct. 
 Magnified 90 diameters. 
 
 ing a branch at its upper end into the centre of a lobule ; the 
 other lobules round about it are evidently cut in a plane that 
 does not show the branches of the duct passing into them. The 
 lobule in involution preserves its entirety ; that is to say, the 
 acini composing it remain as closely packed together in tlie 
 involuted state as in the state of full expansion. There is occa- 
 sionally found a small growth of connective tissue between the 
 
 1 Sitzungsherichte der Wien. Akad., Juli, 1870. 
 
 2 Quoted by Toltlt. 
 
14 
 
 THE PERIODICAL INVOLUTION 
 
 [Pt. I. 
 
 Fig. 4. 
 
 Appearance of an invo- 
 luted lobule with lii^lier 
 magnifying power (300 dia- 
 meters). 
 
 separate acini of a lobule ; this is best seen in the udder of the 
 ewe. It is sometimes also observed that the 
 more outlying acini of a lobule are detached 
 from the rest by a few fat-cells in cases 
 where fat occurs in the gland. But gene- 
 rally speaking the lobule shrinks up en 
 masse, and it is often found that two or 
 three entire lobuli remain closely packed 
 together. 
 
 The involuted acini present very definite 
 characters, which have been found to occur 
 with great uniformity. Each acinus appears 
 on section as a rounded space bounded by 
 a continuous thin membrane, which gene- 
 rally colours deeply with the staining fluid. 
 It most commonly appears as if the lobule 
 were made up of a number of independent 
 round spaces, but the infundibular or trefoil-like appearance of 
 several communicating acini can often be seen. The diameter of 
 an involuted acinus is about one-fourth that of the acinus during 
 lactation. The contents of the involuted acinus are the chief 
 point of interest. Instead of the cellular elements being to the 
 number of 15 to 20 or more, as they are seen to be in the floor 
 of a perfect acinus, they number, in the sectional view, on the 
 average half-a-dozen ; and instead of forming a mosaic of large 
 polyhedric cells of which the central nucleus is not more than 
 one-third of the whole breadth of the cell, they are nothing else 
 than a somewhat irregular heap of naked nuclei, with no fringe 
 of protoplasm round them, and in size little if at all larger than 
 the nucleus alone of the perfect epithelium. The cellular bodies 
 of the involuted or resting state are however not all of them 
 round or elliptical ; a certain proportion of them are more oblong, 
 others are crescentic or more or less triangular. They colour with 
 the staining fluid the same as nuclei ; the more round or oval 
 forms show a nucleolus. 
 
 The involuted acinus is to be regarded as a spherical space 
 more or less completely filled with a number of nuclear bodies. 
 The elements of the basement membrane are not readily made 
 out. There are now and then seen what appear to be sjjindle- 
 
On. I.] 
 
 OF THE BREAST. 
 
 15 
 
 shaped or oblong thickenings in the substance of the limiting 
 membrane ; but it is doubtful whether these are not to be classed 
 with the heap of cells within. In like manner there are seen, 
 among the cells in the interior, stellate and other forms which 
 cannot always be distinguished from the crescentic or other irre- 
 gular forms of cells belonging without doubt to the general intra- 
 acinous group of cells. 
 
 Capillary vessels of a relatively large size can be seen branch- 
 ing in various directions among the acini of the lobule ; but they 
 are not seen to encircle the individual acini in the same manner 
 as in the state of full expansion. 
 
 Having thus far described the fully-expanded state of the 
 secreting structure, and the involuted or resting state of the same, 
 we are now in a position to trace, from the intermediate condi- 
 tions of the gland, the entire course of the involution process. 
 
 The adjoining figure is a low-power view of a lobulus from the 
 mamma of a cat that had suckled its young for sixteen days, and 
 had then been kept apart from them for nine days. When the 
 
 Fig. 5. 
 
 
 ^^i 
 
 From the niainma of a cat nir.e days after laptation. Lolmlo 
 retaining its full expansion, with a change in its epithelium. Blagiiilied 
 90 diameters. 
 
 animal was killed at the end of that time, the mammary glands 
 were found little if at all diminished in volume, and to have the 
 ordinary prancreas-like structure of the lactation period. The 
 substance was, however, not at all loaded with milk. On micro- 
 scopic examination, the lobuli, as the figure shows, were found fur 
 
16 THE PERIODICAL INVOLUTION [Pt. I. 
 
 the most part to be of the same size as in the state of full 
 development, and the connective tissue separating them was 
 inconsiderable. The acini also were of much the same shape and 
 size as in the secreting gland. But their contents were totally 
 different. While they retained their size and form, their floor, 
 instead of being covered with a mosaic of polyhedric epithelium, 
 presented the following appearances, which were characteristic 
 ever3'where of the gland-structure of this animal. It may be 
 stated also that the appearances were the same in various sets 
 of preparations, whether hardened in chromic acid or in methy- 
 lated spirit, and whether coloured with logwood or carmine. 
 Each acinus had strewn over its floor, or arranged at somewhat 
 wide intervals round its margin, a number of deeply-stained 
 cellular bodies of various shapes. Some of them were round, 
 some more or less cubical, a certain number were of a somewhat 
 thick crescentic shape (hemispherical), and others were triangular. 
 Their size corresponded generally to that of an epithelial nucleus, 
 and their substance coloured unifor dy throughout. It was at 
 the same time apparent that the floor or circuit of the acinus was 
 also occupied by a number of delicate ring-like forms of a greyish 
 unstained granular appearance, and that the variously shaped 
 Fift. 6. cellular elements were seated on their peri- 
 
 phery (Fig. 6). This was most obvious in the 
 case of the crescent-shaped coloured masses, for 
 they completed the circuit of which the thin 
 uncoloured thread formed the larger part. The 
 triangular-shaped bodies were also very signifi- 
 ^ , „ ,, cant : they were found situated below the 
 
 Tlie coiitPiits of the . 
 
 acini in Fig. 5 undor point of coutact of two such hoops Or rings, a 
 
 a hifrluT imignifving ^ _ . , . 
 
 power (.ido diameters). scoopcd-out side of the triangle forming a part 
 of each circle, and the single triangular body evidently repre- 
 sented the fusion at their opposed surfaces of two independent 
 crescentic bodies. 
 
 The interpretation to be given of these appearances will be 
 made less doubtful by describing what is found in a somewhat 
 different preparation of which Fig. 7 is a drawing. This is from 
 the udder of a ewe which was said to have been killed a short 
 time after the cessation of the milk. The acini were generally 
 of a large size, and their contents showed several interestino- 
 
Cii. I.] 
 
 OF THE BREAST. 
 
 17 
 
 modifications of tlie condition observed in the preceding case. 
 The sections showed more frequently the profile view of the 
 
 Fig. 7. 
 
 |,^/p^ .. ^V -.J ^/ )i^'bt^ 
 
 From the udder of a ewe shortly after the end of lactation. Vacuolation of 
 the epithelium in situ. Magnified 300 diameters. 
 
 epithelium, in which the cells were arranged as a complete circlet 
 adhering round the circumference of the acinus ; and many acini 
 showed, in addition to the circlet of epithelium in situ, an irre- 
 gular heap of cells lying free in the lumen of the acinus. 
 
 The portion of the gland of which Fig. 7 is an accurate draw- 
 ing did not show the latter condition, but the group in Fig. 8 are 
 representative forms of the cells that are frequently seen to 
 occupy the lumen of the acini in other parts of the preparation. 
 The cells m situ are vacuolated cells, with the usual thin and 
 for the most part uncoloured hoop or ring of the vacuole, and 
 the deeply stained peripheral^ mass. The most definite and 
 unmistakeable form of vacuolation is the signet-ring type, which 
 c. c. 2 
 
18 THE PERIODICAL INVOLUTION [Pt. I. 
 
 t)ccurs over and over again in these preparations. Another 
 ,, not uncommon appearance, which would be 
 
 ambiguous if taken by itself, is where the 
 coloured or nuclear element of the epithe- 
 lium is, to use the current expression, set 
 free from the parent cell, which is evidently 
 not destined to survive the setting free of 
 the young cell. In those cases the definite 
 boundary of a vacuole and the excentric 
 „ , . ,. , position of the coloured mass are not ap- 
 
 Characteristic appparance of -t^ >- 
 
 S^aS?ni'if.'af,triTlSo[ P^^-e^t, but the phenomenon is one of the 
 :.rt"pr..7rMa.nifledToo ^^me ckss. In the groups of cells that 
 diameters. ^j.q fouud lying free in the lumen of 
 
 an acinus, the process of excavation or vacuolation of the cell 
 substance is very obvious. The cells are true hollow spheres, 
 showing the entire cavity, and not merely a section of it. 
 Individual instances may be observed among them in which the 
 cell substance is seen to be simply granular and not vacuolated ; 
 but in these also the excentric position of the nucleus can be 
 seen. Others, again, of the group are round granular cells, show- 
 ing neither nucleus nor vacuole. The size of the various cells, 
 whether of those in situ or of those in the free cavity of the acini, 
 will be found to correspond on the whole to the size of the perfect 
 mammary epithelium. If the latter were to become globular and 
 to retain their greatest breadth as their diameter, they would 
 have much the same size as the cells now described. 
 
 There is thus a gradual transition from the perfect mammary 
 epithelium to the forms that characterise the various stages of the 
 involution process. The two cases that have been described and 
 figured are merely good examples of what is found uniformly to 
 occur. The various characteristic appearances in the epithelium 
 are found to occur throughout a period extending to the fourth or 
 sixth week after lactation. If the cellular elements of the acini in 
 Fig. 6 are again referred to, it will be seen that they are not un- 
 like those described for the completely involuted or resting state. 
 The remaining steps of the involution process as regards the acini 
 may in the meantime be supposed to be the contraction of the 
 space of the acini as figured in Fig. 5, and the retention of a cer- 
 tain part of the cells within them. This account will have to be 
 
Cii. I.] OF THE BREAST. 19 
 
 supplemented later by some important particulars, but it is con- 
 venient to leave it at present as it stands. 
 
 The ultimate contraction of the circuit of the acini evidently 
 depends on the collapse or yielding of the vessels that encompass 
 it, and make up the most substantial part of its framework. This, 
 again, is related to the diminished supply of blood during the 
 subsidence of the secretion. As the acini contract they preserve, 
 as has been already said, the unity of the lobule, and it is not 
 between the several acini but between the several lobules, and 
 sometimes groups of lobules, that a space is set free. There is, 
 as has been explained, an absolute shrinking up of the whole 
 gland. But there is also an increase of the connective tissue 
 between the lobules, which goes on _2ja?'i passu with the retraction 
 of the latter; and where there is no actual increase in the quantity 
 of the interlobular tissue, there is a formation of fat in it. The 
 production of young fat-cells among the interlobular fibrillar 
 tissue may be clearly seen. The increase of the cushion of fat 
 beneath the gland has already been mentioned. The economy 
 of this process of fat-production is obvious. By means of the sub- 
 mammary layer of fat, the contour of the body is preserved, and 
 by means of the fat within the gland the necessary intervals 
 between such permanent structures as the large ducts and blood- 
 vessels are filled up. In both situations there is the greatest 
 amount of space occupied by employing the least number of cellular 
 elements. 
 
 Such then is the process of involution in the epithelium and 
 in the other tissues of the mamma in so far as its morphology is 
 concerned, or in so far as the simple description of its various 
 phases can carry us. It remains now to enter on a discussion of 
 the physiological significance of the process. 
 
 The epithelium of a mammary acinus examined while the 
 animal is in milk is found to be coarsely granular ; but it is not 
 always easy to find good examples of the perfect epithelial cells 
 where the formation of milk is actively going on. The character 
 of the epithelium, as has been already stated, can be best seen in 
 the gland shortly before parturition. It is generally admitted that 
 the milk is formed at the expense of individual epithelial cells, 
 and that the epithelial cell perishes in being transformed into the 
 fluid of the secretion. There is, indeed, an ob.servation of Strieker's, 
 
20 rilK PERTODTCAL IXVOLUTIOX [Ft. T. 
 
 which is taken to show' tliat the same epithelial cell may yield 
 successive contributions to the secretion. He observed that colo- 
 strum cells swimming in fresh milk gave out fat globules, while 
 they retained their form. The observation as far as concerns 
 colostrum cells is no doubt correct ; but it is at the same time clear 
 that the same colostrum cells could not return to their place on 
 the wall of the acinus and resume their function of milk-produc- 
 tion. Whatever properties they may have had as free cells, their 
 career as epithelial cells was over. The cells in Fig. 8 are identical 
 with colostrum cells, and it has been already mentioned that 
 they occurred in the lumen of acini, the walls of which were also 
 covered with a set of epithelium, the colostrum cells being in fact 
 the predecessors of those that then had the epithelial position. It 
 may, indeed, be taken as settled that the secretion is produced by 
 the transformation of the substance of epithelial cells and the 
 necessary destruction of the cells^ This process means no less 
 than the constant renewal of the epithelium ; and here we are 
 met with a difficulty. The mammary epithelium is an epithelium 
 of one layer, and there is no doubt a difficulty in representing to 
 the mind the double process, the secretory process and the forma- 
 tive process on which it depends, as taking place in an epithelium 
 of one layer. The type of the secretion is, however, the same as in 
 the sebaceous glands, and that of the latter may be compared to 
 the desquamation of the neighbouring epidermis. Now, in the 
 case of the skin, the numerous layers of cells enable the process to 
 be clearly seen. In the lowest layers there is a multiplication of 
 cells by the ordinary hyperplastic process of division; a layer of 
 cells that have begun their career in this manner as formative or 
 
 ^ Langer, Strieker's Handhuch, p. 632. 
 
 " The following extract from an article by Professor Voit expresses the now 
 accepted view of the formation of milk, and illustrates that kind of secretion by 
 contrast with another : 
 
 " The cells or cell-derivatives of many parts of the body effect, for an indefinite 
 time and without themselves perishing, the most active decompositions of the 
 substances passing through them; such, for instance, are those of the liver, of 
 muscle, of the central nervous organs. The cells of the blood, the epidermic and 
 epithelial cells, and also those of the breast, are however made subject to change ; 
 and for the latter the theory of secretion remains true that was stated generally by 
 Goodsir and Meckel, according to which the dissolved gland-cells represent the 
 secretion. Milk is not a product of the activity of cells of the gland, but it is the 
 actual cell-mass become ^\\\dLr—(Zeitschrift fiir Biologie, 1869, p. 140.) 
 
L'n.J.] OF THE BREAST. 21 
 
 plastic cells come iu their tui'u to the surface, and undergo the 
 horny transformation, after which they are cast off. There is here 
 no special class of cells which are set apart for formative activity ; 
 the same cells are at one time iu a formative stage and at another 
 time in their functional stage, and the many layers of the cells 
 enable us to see the two incidents of their life apart. The differ- 
 ence in the mamma is that there is visibly present no new genera- 
 tion of cells in readiness to take the place of those that are cast 
 off, but that difference entirely depends on the epithelium being an 
 epithelium of one layer. The same cycle which is, as it were, un- 
 rolled in the rete mucosum and epidermis, is implicitly present in 
 the activity of the mammary epithelium. With this explanation, 
 it will be convenient to separate the notions of the formative and 
 of the functional activity of the epithelial cells, the latter constantly 
 presupposing the former. 
 
 The secretion then may be said to be produced by a transforma- 
 tion of the substance of successive generations of epithelial cells, 
 and in the state of full activity that transformation of the substance 
 is so complete that it may be called a deliquescence. 
 
 The process in the cells presents in fact no morphological 
 characters whatever. But when the secretory activity of the gland 
 is weakened, as in the subsidence of the function during involu- 
 tion, the morphological characters of the process begin to become 
 apparent. Reverting to the appearances of the early stage of in- 
 volution, as in Fig. 7, it will be seen that the process in the 
 epithelium is one of vacuolation. The substance of the cell is 
 transformed into a cavity filled with a fluid, and a certain portion 
 of the cell's substance remains as the wall of this cavity and as a 
 peripheral mass, which colours like the original nucleus. Now, 
 among vacuolated epithelial cells, there will be observed consider- 
 able differences in the bulk of the peripheral mass and in the 
 thickness of the wall of the vacuole. In some cases, the former is 
 a large cellular element, as in some of the cells of Fig. 6 ; in others 
 it is a thin crescent-shaped line, which is distinguishable from the 
 rest of the vacuole outline only by its property of staining. Cells 
 of this variety may be seen in situ in Fig. 7. We are here in the 
 midst of a series of gradations. In the vacuolation process there is 
 evidently a correlation between the amount of the cell-substance 
 that is devoted to the vacuolar or fluid product, and the amoimt 
 
22 Tin: PERIODICAL IX VOLUTION [Ft. I. 
 
 that is devoted to the cellidar product. One extreme of this series 
 is where the vacuole reaches the maximum, leaving the peripheral 
 or cellular mass at its lowest possible amount ; and the other 
 extreme is where the vacuolar product is at its minimum, and the 
 cellular at its maximum. Both of those extremes have an exist- 
 ence in fact; with the latter we shall have nothing to do at present, 
 but the former is simply the perfect transformation of the mam- 
 mary epithelium into milk, which has been called a deliquescence. 
 Now there is evidently not one law for the function at its height 
 and another law for the function in its subsidence ; there is a per- 
 fect continuity between the full secretory activity of the gland in 
 the lactation period, and the diminished secretory activity which 
 carries the gland through the involution process to the resting 
 state. The functional stimulus, whatever it is, is simply lowered 
 in the course of involution; it is not that the secretory force 
 abruptly ends when lactation ends, and that the subsequent changes 
 in the secreting jjarenchyma are of a merely retrograde or degenera- 
 tive kind. 
 
 The perfect transformation of the mammary epithelium into 
 milk is therefore, in its essence, a process of complete vacuolation. 
 But the significance of that fact is missed unless the vacuolation 
 process be reduced to its proper place among the great cellular 
 processes of the organism. If the vacuolation process be examined 
 in those cases where its morphological features are most declared, 
 it will not perhaps seem fanciful to describe it as an example of 
 endogenous cell-formation. 
 
 Less has been said of endogenous cell-formation in recent 
 writings than in the earlier days of the cell doctrine. The two 
 kinds of cell- formation — that by fission or division and that by the 
 endogenous mode — are less rigidly contrasted than formerly; in 
 botanical works, especially, there are other and intermediate ways 
 of cell-renewal and cell-multiplication enumerated. But the endo- 
 genous mode retains its place in the text-books of animal histo- 
 logy ; and if there were any doubt as to its occurrence in the 
 healthy body, an examination of certain pathological processes 
 would at once establish its distinctness and its contrast to the more 
 usual mode of cell-formation by fission or division. In a former 
 paper, on the histogenesis of secondary tumours in the liver, I laid 
 jjreat stress on the part that the endogenous mode of cell-forma- 
 
Cn. I.] OF THE BREAST. 23 
 
 tioa played in producing the tumour cells out of liver-cells. In 
 Chapter VIII. I shall describe the same process as occurring still 
 more distinctly in lymphatic glands infected from tumours ; it has 
 been often described by other writers as occurring in the lymphatic 
 glands in almost every infective disorder. A study of the various 
 forms of cells undergoing the endogenous process in the liver led 
 me, in my first paper, to compare them with the vacuolated cells 
 of other pathological conditions, and with the vacuolated cells that 
 are found in certain healthy processes which I enumerated. These 
 resemblances seemed to me to point to a theory of secretion, and 
 I concluded that the formation of secretion in epithelial cells was 
 " essentially a process of endogenous cell-formation." It would 
 require a large collection of facts, gathered especially from com- 
 parative physiology, to show some probability for this hypothesis 
 as applied to all secreting cells, and I now limit my statements to 
 the secretion of the breast. 
 
 When this view of secretion first occurred to me, I searched for 
 some earlier statement to the same effect, and in an article by 
 Professor Virchow, written in 1852, I found a quotation from 
 another writer that appeared to be directly to the point, and which 
 I referred to at the time as supporting the view I had inde- 
 pendently taken up. According to Virchow, it was maintained by 
 Friedrich Will that " all secretions proper are produced by means 
 of cell-formation, and in fact by means of endogenous cell-forma- 
 tion {durch Zellenhildung, und ziuar durch endogene Zellenbildung 
 vermittelt).'' The first part of the statement is nothing more than the 
 generalisation that Goodsir first clearly laid down. But the second 
 part of the statement, although it might be inferred from many of 
 the facts that Goodsir collected, is at variance with his conclu- 
 sion that "growth and secretion are identical," and must be re- 
 garded as a distinct addition to the cellular theory of secretion. In 
 making the above quotation from Will, Virchow adds : " It is to 
 be hoped that the proposition thus broadly stated is to be taken as 
 applying only to the secretions of invertebrate animals ; but it 
 seems, at any rate, to indicate the wide distribution of a pheno- 
 menon that has perhaps an increasing significance in physiology, 
 and which may lead us in course of time to regard the process of 
 secretion from a higher and more general point of view\" 
 
 ^ Yirchow's Archie, iii. 221. 
 
24 THE PERIODICAL IX VOLUTION' [Pr. I. 
 
 That surmise cau hardly be said to have proved correct. In so 
 far as modern text-books of physiology touch upon the cellular 
 side of secretion, they give a merely descriptive or pictorial ac- 
 count of it. The following example, quoted from a work mainly 
 pathological, may be taken as showing how much progress has 
 been made in the cellular theory of secretion^ : " C'est ainsi que 
 dans les glandes de I'estomac on voit des cellules primitivement 
 cylindriques devenir spheriques en se gorgeant de sues, tomber 
 dans la lumiere de la glande et se detruire en laissant echapper 
 leur contenu." A more complete chapter of accidents, a more 
 gratuitous tragedy, could hardly be pictured. But if precisely the 
 same series of events be interpreted by the law of endogenous cell- 
 formation, and if that mode of cellular activity be viewed in con- 
 trast with the other great variety. Viz. by fission or division, we 
 get an account of the process of secretion which may be called 
 rational as compared with the one just quoted. What distin- 
 guishes the endogenous kind of cell-formation is the production 
 along with the young cell of a vesicle or vacuole containing a 
 fluid ; in cell-formation by division the parent cell simply breaks 
 up into two or more young cells. One may say that in the endo- 
 genous variety there is a provision for a fluid product along with 
 the solid or cellular; and it is according to many analogies in nature 
 of the adaptation of a general plan to particular ends that such 
 provision should be magnified into the essential feature of the 
 process. Whether it is in this sense or in some other that Will 
 speaks of secretions as being produced by means of endogenous 
 cell-formation, I have been unable to find out, not having had 
 access to his original paper. 
 
 The foregoing paragraph is for the most part a repetition of 
 what I wrote in the paper already quoted ^ The more extended 
 study of the secretion of the breast, in particular the study of it 
 throughout its periods of subsidence and restoration, enables me to 
 maintain this view of the mammary secretion with greater con- 
 fidence. Not only is vacuolation seen to be the true morphological 
 change in the secreting cell, but, as the function subsides, there is 
 found an actual and even abundant production of cells from the 
 epithelium, within or along with the fluid-containing vacuole; and 
 
 ' Manuel cC Histologic Pathologiquc by Con il aiul Eaiivier, p. 35. 
 - licpurt of the Medical 0£kcr of the Privy Council for 1871, p. IKL 
 
Cir. I.] OF THE BREAST. 25 
 
 the weaker the function becomes, the greater is its celkilar pro- 
 duct. It is in the middle period of the subsidence that the endo- 
 genous kind of cell-activity is unambiguous. But from knowing 
 the mode of action in the intermediate terms of what is evidently 
 a continuous series, we are enabled to determine what is essen- 
 tially the process in the extreme terms of the series. The expres- 
 sion that takes in all the phases of activity in the mamma, from 
 the perfect secretion through the subsiding period to the resting 
 state, is to call the process a process of endogenous cell-formation. 
 What may be called the law of the secretion is the law of endo- 
 genous cell-formation. 
 
 Having finished the discussion of this more or less abstract 
 but by no means unimportant point, it will now be convenient 
 to introduce certain additional facts which have been omitted 
 in the description of the involution process. These relate to 
 the cellular products of the subsiding secretion of which mention 
 has just been made. They will be frequently mentioned in the 
 sequel under the name of waste products : in the physiological 
 chapters the various normal circumstances under which they 
 are produced and the manner of their safe disposal or u.tilisation 
 in the organism will be fully described ; in the pathological 
 chapters their excessive production, or their production out of 
 due season, or their retention at or near their place of origin 
 will be found to be among the chief factors in the causation 
 of tumours of the breast, and the mode of their production by 
 endogenous cell-formation will be made to furnish a theory of 
 their heteroplastic nature or of the malignancy of the growths 
 which they help to build up. 
 
 Fig. 9 represents a section through the corner or angle of a 
 mammary lobule. The preparation was made from a ewe that 
 was killed three weeks after giving birth to a dead lamb and 
 that had not been milked. At that period, three weeks after 
 parturition, the udder was still of considerable size, and a clear 
 brownish fluid could be expressed from the teats. The microscopic 
 examination showed in many points a resemblance to the con- 
 dition of which Fig. 5 is a drawing; there was a corresponding 
 alteration of the epithelium, and many acini contained a fluid 
 which coloured purple with the logwood and was on that account 
 to be taken to be of a mucous nature. But the distinguishing 
 
26 
 
 THE PERIODICAL INVOLUTION 
 
 [Pt. I. 
 
 feature of the case was the enormous number of lymphoid cells 
 which were met with both in the acini, in the spaces immediately 
 outside them (known to be lymphatic spaces) ', and in the inter- 
 lobular fibrillar tissue. 
 
 Fig. 9. 
 
 
 From the uddtr of a ewe that gave birth to a dead lamb and was not milked. The condition 
 three weeks after parturition: a group of seven acini forming the corner or angle of a lobule; 
 many small round cells in the spaces round the acini and in the fibrillar tissue; mucus, mixed 
 with small round cells, in the cavities of the acini. Magnilied 150 diameters. 
 
 The section represented in Fig. 9 passes through the extreme 
 pointed end of a lobule, and includes only seven acini ; the inter- 
 lobular fibrillar tissue is for the same reason cut obliquely, and 
 appears thicker than it really is. The fibrillar tissue will be 
 seen to be crowded at certain points with lymphoid cells, often 
 ranged as if in a procession. The spaces round the acini, Avhich 
 in the active secretory state are reduced to a mere line, are here 
 
 1 Coine (Soc. de Biologie, 21 Nov. 1871). 
 
Ch. I.] OF THE BREAST. 27 
 
 seen to be broad tracts crowded with the same lymphoid cells ; 
 it is not imfreqnently found in the preparations of this case that 
 the contour of the acinus is completely obscured by the number 
 of small round cells about it. The cells are for the most part 
 round nuclear bodies, but there are found among them the various 
 shapes given in Fig. 6. On closer examination, it is not difficult 
 to find everywhere traces of the vacuolation or endogenous process 
 within the acini, and it is hardly to be doubted that all of those 
 small round cells are derived from the epithelium by that process. 
 The case is one in which the cellular product of the vacuolation 
 or endogenous process is the predominant, and it will be noticed 
 that the accompanying fluid product is not the fatty fluid of 
 the perfect secretion, but a mucous or albuminoid fluid*. The 
 case shows a production of cellular elements from the epithelium 
 which is excessive for that stage of involution, and the fact of 
 the animal never having been milked is perhaps the modifying 
 circumstance ; but the same phenomenon is found to occur in 
 other cases in corresponding stages of the involution process. It 
 will be observed that the greater number of the cells have escaped 
 from the acini into the extra-acinous spaces and the fibrillar 
 tissue, and it is only in those situations that they can be observed 
 in any quantity ; but it is probable that the ordinary outlet of 
 the ducts, which are still patent in that stage of involution, 
 must give exit to many of them along with the fluid that is 
 produced at the same time, and which was found in the ducts. 
 Reverting to the case from which Fig. 5 is taken, the same ap- 
 pearance is seen to a more limited extent. There are here and 
 tliere in the tissue outside the acini considerable heaps of lym- 
 phoid cells ; and among them are found other and thoroughly 
 well-marked waste products of the involution process, which 
 enable us to place the subject on which we are now entering 
 on a perfectly sound basis of evidence. 
 
 The account given provisionally of the completion of the 
 involution process was that the acini, denuded of their perfect 
 mosaic of epithelium, gradually closed upon the nuclear elements 
 that survived from the vacuolation of the latter, and so con- 
 ducted the gland to the resting state as figured in Fig. 4. If 
 
 J " After the secretion of milk has ceased, the secretory structure is often londed 
 with mucus." — Astley Cooper, Anatomij of the Bremt. 
 
28 TIIB PERIODICAL IN VOLCTIOX [Pt. I. 
 
 that were the entire process, we should still have to account 
 for tlie escape or discharge of a certain quantity of cells from 
 the acini; for the resting cells of an acinus are mucli fewer in 
 number than those that cover its floor in the middle stage of 
 involution. For example, in the case from which Fig. 5 is taken, 
 there are' found considerable tracts of acini which have already 
 contracted their circumference to some extent, and there is to 
 be seen at one or two places a lobule which has for some reason 
 or other advanced beyond the others, and reached the extreme 
 contracted state of involution. In such a lobule the cells within 
 the acini are extremely few, from four to six in the section. 
 But what arrests the attention at once in looking at this lobule 
 is the presence in it of a quantity of bright yellow pigment. 
 On closer inspection the pigment is seen to be for the most 
 part scattered about the cells inside the acini; and if we leave 
 this involuted lobule and search in less advanced parts of the 
 gland, the source of the pigment and the significance of it become 
 clear enough. The other class of waste products which were 
 referred to above as being thoroughly well marked and as 
 affording a sound basis of evidence for the entire question of 
 epithelial waste products, are large granular nucleated cells filled 
 with a bright yellow or golden pigment. The nucleus is often 
 visible ; and when it is stained purple with logwood, the cell 
 with its purple nucleus and its bright yellow substance becomes 
 to the microscopist a striking and pleasing object. Fig. 10 
 represents a very common appearance of the large yellow cells. 
 The lobule is in a somewhat contracted state ; in several of the 
 acini the pigmented cells, which are represented in the woodcut 
 simply as granular, may be seen in situ; while a considerable 
 number of them are seen to occupy the interfibrillar spaces 
 outside the acini and the lobule. The latter situation is most 
 characteristic of them. The figure is not taken from a gland 
 known to be in the process of involution ; on the other hand 
 it probably represents an early stage of evolution. But it is 
 inserted here to show once for all the large yellow granular cells 
 to the best advantage as regards their size, shape and general 
 appearance, so far as a woodcut can reproduce them. 
 
 Out of a great many preparations of the involution process 
 it would be difficult to find one in which the origin and destiny 
 
Ch. I] OF THE BREAST. 29 
 
 Fig. 10. 
 
 W^S^! 
 
 
 Lobule of a mamma near the resting state. Numerous larpre pigmented cells within the 
 acini and in the interlobular fibrillar tissue; sections of three blood-vessels to the left of 
 the lobule. Magnified 160 diameters. 
 
 of the pigmented cells together with their place in the involution 
 process were at once apparent. They are not unfrequently seen in 
 the tissue outside a lobule in rows three or four deep ; again they 
 are found in the interfascicular spaces among the lymphoid 
 cells that have been already mentioned. Inside an acinus they 
 are found seldom in greater numbers than one or two, but in 
 that position they do not always possess the definite characters 
 that they often show elsewhere. What one sees inside an acinus 
 is, as it were, certain round discs or other irregular forms of 
 pigmented substance generally bestowed round a nucleus ; this 
 is a common appearance in the preparation from w^hich Fig. 5 
 is taken. Finally, as has been already said, the same j)reparations 
 show, in the involuted lobules that they contain, a more or less 
 amorphous appearance of the pigment within the acini, and in sets 
 of preparations from other animals (especially the rabbit), showing 
 the advanced involuted state, the surface of the involuted lobules 
 is found to be strewn with numerous grains of pigment of a 
 brownish colour. Putting: together these facts collected from 
 the involution process alone, a theory of the origin of the pig- 
 mented cells might be made out. But the question of their 
 origin and significance is greatly simplified by anticipating so 
 far the study of the evolution process and of certain pathological 
 
30 
 
 THE PERI ODICA L IN \ 'OL ( 'TIOX 
 
 [Pt. T. 
 
 processes. A drawing of one of the latter may liere be inserted. 
 Fig. 11 is the section of a pathological lobidus from the bitch: 
 
 Fig. 11. 
 
 m 
 
 m 
 
 J) 
 
 \i 
 \ 
 
 \ 
 
 fT«w' 
 
 
 .V"'^, 
 
 
 ^ 
 
 
 I'^i'^'" "' 
 
 A pathological lobule. The uppermost acinus shows the pigmented 
 cells as epithelium in situ. Magnified 160 diameters. 
 
 the circumstances of its occurrence do not require to be stated 
 here. It is found among lobuli most of which show the ordinary 
 appearance of the resting mamma ; and this particular lobulus, 
 though three or four times the size of its neiohbours, has retained 
 its proportions exactly. All the acini of this lobule have a 
 yellowish tint; but the attention is chiefly directed to the half 
 acinus that is uppermost in the figure. In it the cells are yellow 
 and granular, and most of them show a nucleus; they are also 
 for the most part columnar or club-shaped. They occupy the 
 position of the normal mammary epithelium, and the pathological 
 form that they have acquired makes them, if one may so s})eak. 
 
Ch. I.] OF THE BREAST. 31 
 
 more epithelial than the mammary epithelium usually is. Yet 
 they look as if they were about to be shed from their epithelial 
 situation, and as a matter of fact the same cells, with trifling- 
 modifications of form, are to be seen in considerable numbers 
 in the adjoining fibrillar tissue. In the pathological part of 
 the work many other instances will be given of this peculiar 
 transformation of the epithelium ; what is here desired is simply 
 to show the epithelial origin of the pigment-cells of the mamma. 
 So conclusive are all the appearances bearing on this question, 
 that it • may be stated in the broadest manner that, whenever 
 those large yellow granular cells are found in the mamma, they 
 are of epithelial origin. 
 
 Havinsr so far stated the evidence for the uniform occurrence 
 of the large granular pigmented cells in the course of the 
 periodical involution, and having described their more general 
 characters, we come now to consider what is the exact place 
 that is to be assigned to them in the involution process, and 
 by what considerations they may be brought Avithin the operation 
 of the general law of the secretion, that is to say, the law of 
 endogenous cell-formation. 
 
 As the large colostrum-like cells that are figured in Fig. 8 
 characterise an early stage of involution, and as the more dis- 
 ' tinctly nuclear products of vacuolation of Figs. 6 and 9, together 
 with the accompanying mucous fluid, characterise the subsequent 
 stages, so the pigmented cells are found to be characteristic of 
 the last state of involution, and the pigment that belongs to 
 them is to be found strewn over the lobules that have reached 
 the resting state. It has up to this point seemed a satisfactory 
 account of the involution process to say that the acini contracted 
 their space and retained within them a certain number of the 
 cellular bodies that remained over from the vacuolation of the 
 epithelium. But the unexpected introduction of a set of pig- 
 ment-bearing cells is precisely one of those matters of observation 
 wherein the actual course of nature proves more varied than 
 had been conceived, and whereby we are led to enlarge very 
 considerably our view of the process. 
 
 In reviewing the process of involution in the mamma, it 
 might appear at first sight as if the entire cycle of changes in 
 the epithelium, that conduct the gland to the resting state, 
 
32 THE PERIODICAL IX VOLUTION [Pt. I. 
 
 .took place iu one and the same set of cells. When lactation 
 stops, the acini may be imagined to possess a mosaic of perfect 
 epithelial cells which have not followed their predecessors in 
 transforming- their substance into milk, but which are destined 
 to be the elements that the resting cells of the gland are, after 
 a prolonged process, formed out of. But a cursory view of the 
 facts will at once show that such is not the case, A preparation 
 has been already described, belonging to an early stage of in- 
 volution, in which many of the acini had a complete set of 
 vacuolated cells lining their wall, and another collection of similar 
 cells lying free in their lumen ; and this fact can mean nothing 
 else than that the cells now in the cavity of the acinus had been 
 shed from their epithelial situation, and had been replaced by 
 another set. In like manner the various forms of cells that 
 characterise the various stages of involution must have resulted 
 from a transformation de novo of the renewed epithelium, and 
 not from successive changes upon the same cell. Thus, it is 
 difficult to imagine the peculiar pigment-transformation of the 
 epithelium to be imything else than the direct and sole change 
 that those particular cells undergo. They are evidently products 
 sui generis ; they are the kind of cells proper to the later stages 
 of involution ; and it may be asserted, by way of illustration, that 
 if the functional stimulus of the gland always remained at the 
 same low intensity that it has in that stage of subsidence, the 
 gland would produce nothing but those cells. The notion then 
 must be set aside, that the particular set of cells that the gland 
 possesses when lactation stops, are the same which become, after 
 various transformations, the resting cells at the end of involution. 
 The epithelium, which had been constantly renewed during 
 lactation, is renewed also many times before the involution is 
 completed, and before the functional activity subsides. In other 
 words, the formative or plastic activity of the gland goes on 
 so long as its functional activity goes on. As the latter gets 
 weaker, the foi'mer gets slower, till a point is reached when there 
 is no further renewal of cells within the acinus, and those last 
 produced remain almost without change. 
 
 Thus, in concluding the descriptive account of the process 
 of involution in the mamma, we are left with the important 
 physiological fact that the periodical subsidence of the function 
 
Ch. I.] OF THE BREAST. 33 
 
 is accompanied by a considerable production of cellular waste 
 material. In the earlier stages of involution it is thrown off in 
 the form of the cells that are drawn in Fig. 8, and that class of 
 waste products doubtless find their way out of the gland by 
 the ducts. The cells in Fig. 9, along with the variously shaped 
 peripheral masses of the vacuoles, may be taken to represent 
 the waste products of a later stage, and it is seen that the 
 lymphatic circulation is at least one mode of their disposal. 
 Lastly, towards the end of the process there are produced the 
 very remarkable yellow cells, and these also find their way out 
 of the gland by the lymphatic channels. The complete de- 
 scription of the removal of the cellular waste of the secretion 
 by means of the lymphatic circulation is postponed till the evo- 
 lution process has been described. 
 
 There remain for discussion, in connexion with the process 
 of involution, two points of a theoretical nature. One of these 
 will deal with the considerations by means of which the pheno- 
 menon of the large granulaj^-jS^Snted, cells can be brought 
 within the law of the se^^w^VnTCli li'heId^k> l^the law of 
 endogenous cell-formatio»';^ne oR^^P^bMSQaay ^ ^c-iefly dis- 
 posed of at once. Ij 1 ^ |] 
 
 In the state of activel^eecretioH'OT'thejimwHHa, it is/fonvenient 
 to suppose that each ep^^^^ cell that is_ u^^up in the 
 formation of the milk has he^ jMj^&^m^J^r^ii ^ieci polyhedric 
 cell of a homogeneous or finely granular protoplasm and with 
 a central nucleus, and that it has rapidly undergone the cycle 
 of changes whereby its whole substance has been converted 
 into milk. But it is not necessary to suppose that the same 
 should be the case when the function is not at its full force, 
 and is in fact slowly subsiding. The plastic or formative activity 
 of the epithelium must continue, as has been argued, so long 
 as the functional activity has not entirely come to a standstill. 
 But it is not at all probable that the same perfect mosaic of 
 polyhedric epithelium is reproduced at every successive step 
 throughout the whole period of involution as well. For example, 
 if the plastic activity of the gland be viewed for a moment, apart 
 from the functional, and if a point of time be taken, say towards 
 the end of involution, it is not to be supposed that the acini 
 of the gland, at that point of time, would be clothed with a 
 c. c. 3 
 
34 THE PERIODICAL INVOLUTION [Pt. T. 
 
 layer of perfect polyhedric epithelium. On the contrary, the 
 nearer is the approach to complete cessation of the gland's activity, 
 the more crude, so to speak, are the successive renewals of the 
 epithelium. The importance of this theoretical statement will 
 appear when the process of evolution is described. 
 
 The other theoretical point reserved for consideration relates 
 to the pigmented waste-cells. 
 
 As we were led to view the ordinary transformation of the epi- 
 thelium into milk as a process of extreme vacuolation and to refer 
 it to a place under the law of endogenous cell-formation, so we 
 come, at the other end of the scale, to refer the phenomenon of 
 the granular pigmented cells to the same law. The superficial 
 objection to calling the production of milk a process of endogenous 
 cell-formation is that there is in it no actual formation of cells. 
 The objection to classing the granular pigmented cells among the 
 same phenomena is that there is in them nothing endogenous or 
 vacuolar. The explanation of those apparent anomalies is furnished 
 by a method of proof that has already been put in force. Directing 
 attention again to the well-defined vacuolation process in the cells, 
 which marks the middle stages of involution, it will be seen that 
 various degrees of vacuolation of a cell occur, and that there is a 
 correlation between the extent of the vacuole and the extent of 
 the associated cellular mass which is generally on the periphery. 
 It often happens that the vacuole is large, occupying almost the 
 whole substance of the cell, and that the solid part of the cell, that 
 part of its substance which gives it an individual existence and a 
 shape, is nothing more than a delicate ring (in profile), with a 
 slight crescentic thickening of coloured substance at one side. It 
 is by observing the varieties of this peripheral coloured element of 
 the vacuolation that the significance of the process becomes clear. 
 Instead of being a thin rib-shaped or crescentic body, it is often of 
 a rhombic or of a hemispherical form, while the latter forms shade 
 off gradually into the round nuclear bodies which are sometimes 
 found lying as if on the ring of the vacuole, but often also found 
 as free cells in its centre. Again, cells are now and then found 
 that have both the crescent-shaped peripheral mass and the free 
 round cell in the vacuole, and this may be considered as the most 
 comprehensive type of vacuolation. Here, then, the process is evi- 
 dently the process of cell-formation by the endogenous mode; and 
 
Cii. I.] OF THE BREAST. 35 
 
 if the vacuoles of secreting cells were only found to contain several 
 small round cells in their cavity, instead of one, the most rigid and 
 inelastic conception of endogenous cell-formation would be com- 
 pletely satisfied. But the plurality of free round cells in the 
 vacuole, although it is often lield to be an essential part of the 
 definition of endogenous cell-formation, seems to be oftener a 
 pathological phenomenon than a physiological. Regarding, then, 
 the epithelial cell which is subjected to this process as containing 
 a fixed quantity of substance, the more of it that is devoted to 
 fluid transformation, the less is the amount of the co-ordinate 
 cellular product, and vice versa. There is a vanishing point for each 
 of the two factors in the process respectively. The one vanishing 
 point is where everything is fluid, the other vanishing point is 
 where everything is solid; and yet the law of the cell's activity is 
 the same in those extremes as in the intermediate degrees. BotTi 
 of the vanishing points have an existence in fact. The perfect 
 formation of milk is the one, and the purely solid product of the 
 epithelium is the other; the one coincides with the full force of 
 the functional excitation, the other with its extreme enfeeble- 
 ment^ 
 
 There are several matters of observation in connexion with the 
 granular pigmented cells that require to be stated, in order to 
 redeem the somewhat theoretical, and, it may still appear, fanciful 
 account of the conclusion of the involution process. In the mamma 
 of a cat that was killed four weeks after the young were taken 
 away, the acini vvere found considerably reduced in size; they con- 
 
 1 Goodsir, in his Memoir ou Secreting Structures [Anatomical Memoirs, 
 Vol. II. p. 422), drew many of his conclusions from glands in the lower animals, 
 such as the testicle, that have periodical seasons of activity. One of his con- 
 clusions is as follows. ' ' There are three orders of secretions : (1) A true secretion 
 —that is, matter formed in the primary secreting cell-cavities; or (2) A mixture of 
 fluid formed in these cell-cavities with the developed or undeveloped nuclei of the 
 cells themselves ; and (3) it may be a number of secondary ceUs passing out entire." 
 A liver foUicle in a crustacean showed the three orders of secretion at one and the 
 same time. " The blind extremity of the follicle is slightly pointed, and contains 
 in its interior a mass of perfectly transparent nucleated cells. From the blind 
 extremity downwards these cells appear in progressive states of development. At 
 first they are mere primitive nucleated cells; further on they contain young cells; 
 and beyond this they assume the characters of primary secreting cells, being 
 distended with yellow bile, in which float oil-globules, the oil in some instances 
 occupying the whole cell." 
 
 3—2 
 
36 THE PERIODICAL INVOLUTION [Pt. I. 
 
 taiued, very uniformly, ring-shaped vacuolated cells, but they had 
 everywhere as well a yellowish colour, and the yellow tint did not 
 extend to the other tissues of the gland. On closer examination it 
 seemed that the yellow colour was in the fluid contents of the 
 vacuoles, and there appeared to be traces of the yellow granular 
 cells in the fibrillar tissue outside the acini, in the same position as 
 they are clearly seen to have in other preparations better adapted 
 for the purpose. Unfortunately the appearances in this case were 
 always somewhat indefinite, although several sets of preparations 
 were made from it. But to supplement that defective evidence, it 
 may be stated, that in a remarkable case, to be mentioned hereafter, 
 there was found distinct vacuolation of the pigmented cells outside 
 the acini, the fluid contents of the vacuole being distinctly yellow. 
 Again, in the case of a cat nine days after the end of lactation, 
 which has been used for illustration several times already, the 
 yellow granular substance within the acini seemed to break up 
 into a number of round disc-like bodies. It is not desired to follow 
 this subject at present further than to recall the fact that the pig- 
 ment of the granular cells is not unfrequently seen strewn over 
 the involuted lobules, having the appearance of amorphous grains 
 of haemoglobin. The resting cells of the acini, or at least some of 
 them, are in all probability the surviving nuclei of the granular 
 pigmented cells ; the latter have given up their pigmented cell- 
 substance, but their nuclei remain. These facts point to a linger- 
 ing secretory activity in the epithelium that is transformed into 
 pigmented cells. But the greater part of the large yellow cells are 
 thrown off" from the acini entire, and find their way out of the 
 gland entire. Their functional importance is, as it were, so slight, 
 that they are actually thrown off from the gland before any function 
 is performed ; and this consideration leads me to speak, in the 
 briefest possible manner, and as a conclusion to the first part of 
 the inquiry, of the significance of the solid or cellular products of 
 the epithelium as regards the function. 
 
 From the functional point of view, or from the standpoint of 
 usefulness, the fluid product of the epithelium is the only one of 
 the two that is properly speaking a secretory product. But the 
 other or cellular product is all the while potentially present; and 
 it is precisely when the fluid product ceases to be useful or to be 
 required, that the cellular product asserts itself. This cellular 
 
Ch. I.] OF THE BREAST. 37 
 
 product, so far as the facts up to this point carry us, does not 
 become itself useful in its turn; it is indeed spoken of as a waste 
 product. But the formation of the useful product cannot come to 
 an end without the temporary formation of a quantity of waste 
 material. The periodical subsidence of its function costs the 
 mamma a good deal of useless cell-formation, and we shall have 
 occasion to show immediately that the re-establishment of its 
 function, extending over a longer period, costs it a good deal 
 more; while the waste cells produced in both of those processes 
 (and in all other enfeebled conditions of the secretion) have it 
 in their power to become, and sometimes do become, highly dan- 
 uerous elements of disease. 
 
CHAPTER II. 
 
 THE PERIODICAL EVOLUTION OF THE BREAST. 
 
 It has been already explained that the term " evolution " of the 
 mamma is used to describe the periodical process by which the 
 gland, after each period of functional rest, is step by step rehabili- 
 tated with its full structure, iu readiness for the next period of 
 suckling. In an animal killed a day or two before parturition 
 the mamma is found to possess fully expanded lobules and acini, 
 and the acini to have the mosaic of perfect epithelium. Prepara- 
 tions of the mamma at this period are in fact better fitted to 
 show the perfect form of the epithelium than preparations taken 
 from a gland in the full course of lactation. The beginning of the 
 evolution process must be held to coincide more or less exactly 
 with the beginning of pregnancy. Important changes in the 
 gland are observed at early periods of pregnancy, and the evolution 
 is found to proceed, in its later stages, so exactly hand in hand 
 with the growth of the embryo, that it is convenient to suppose 
 the first step towards the preparation of the gland to be contempo- 
 raneous with the early changes in the uterus, although, as a matter 
 of observation, little or no alteration of the resting state is found 
 till the embryo is somewhat grown. We thus arrive at the impor- 
 tant fact, that the evolution process is an exceedingly slow process, 
 in respect of the object to be attained, and that it occupies, gene- 
 rally speaking, the entire period of gestation. 
 
 The evolution process, as about to be described, has been 
 studied chiefly in the cat. Casual preparations have also been made 
 from the sheep and guinea-pig, and a large amount of confirmatory 
 
Cii. ir.] PERIODICAL EVOLUTION OF THE BREAST. 39 
 
 evidence has been obtained from a series of cases in the dog, which 
 were made use of under the following circumstances. A number 
 of cases of mammary tumours in the bitch were collected for the 
 purpose of pathological study ; the tumours for the most part 
 affected only a limited area of the chain of glands on one side. 
 In several of the cases the entire chain of glands was obtained, the 
 animals having been killed in anticipation of a fatal termination 
 of the disease. It was found, on carefully examining the other 
 cases of tumours, which were removed by operation, that nearly 
 all of them had distinct fringes of mammary tissue around them, 
 which, by the side of the tumour, was to be accounted normal. 
 Those cases, fifteen in all, certainly did not, so far as was known 
 of the history of the animals, exemplify the physiological pro- 
 cesses of evolution or of involution. But the microscopic exami- 
 nation showed that they were each of them the subject of a certain 
 feeble kind of activity, which often resembled that of the earlier 
 stages of the evolution process, although the excitation to it was 
 not the same. They seemed to show, in fact, that whatever was 
 the stimulus that stirred up the dormant activity of the gland, it 
 commenced in the track of the evolution process, and that it 
 rarely got beyond the earlier stages of that process, but continued 
 to act upon the gland, so long as it lasted, at a uniform low degree 
 of force. As regards the origin of the tumours, sometimes mul- 
 tiple, that occurred in portions of the chain of glands, this feeble 
 activity was of the highest significance, while the cases were at 
 the same time found to illustrate those effects of weak functional 
 stimulus that are found in the physiological processes, and to 
 coincide rather with the commencing stimulus of the evolution 
 process than with the subsiding stimulus of involution. The series 
 of cases in the cat on which the account of normal evolution is 
 mainly based were eleven in number ; two of these were known 
 to be cases of jpriwi/jjarce. They were fairly representative of 
 various stages of pregnancy, as estimated by the size of embryo. 
 The newborn kitten may be taken to be about 5 inches long, 
 measured to the root of the tail. The youngest litter were 1^ inches 
 long, and the intermediate sizes of 2i, 3, 3J, and 4 inches, were 
 found, as well as foetuses almost full grown. There was on the 
 whole a remarkable correspondence throughout the various cases 
 between the size of the foetus and the staije of evolution of the 
 
40 THE PERIODICAL EVOLUTION [W. I. 
 
 mamma in the pareut. Thus in two sets ot" I'cetuses that were 
 nearly 5 inches long and with the hair grown, the mammse of the 
 parents showed the most minute resemblance, and they both re- 
 sembled the condition in the udder of a ewe that was pregnant 
 with a foetus about 10 to 12 inches long. Again, the mamma? 
 corresponding to the foetuses 3 to 4 inches long had all an equally 
 characteristic appearance. Greatest diversity occurred in the earlier 
 stages of pregnancy. Thus, in a cat with foetuses 2^ inches long 
 the manmiae were found in a condition not further removed from 
 the resting state than in a cat with foetuses of \^ inches, and con- 
 siderably less advanced in- evolution than in a third cat that had 
 also 2^ inch foetuses. 
 
 The three cases last mentioned, taken together, represent the 
 earliest appreciable progress towards unfolding, and the condition 
 found in them will now be described. In naked-eye characters 
 the glands differed little from those in the resting state. Their 
 substance was of no great thickness, tough on section, and with 
 the gland-tissue proper separated into islands or bands by tracts of 
 connective tissue, with or without fat ; the larger ducts could be 
 seen radiating towards the nipples, and the open mouths of them 
 and of large vessels were conspicuous in cross sections. On micro- 
 scopic examination the acini had already advanced from the resting 
 state both in size and in other particulars. In point of size they 
 were perhaps half that of the perfect acinus ; the infundibular or 
 trefoil-like appearance of three or more acini communicating was 
 much more frequently observed in these preparations than in 
 those of any other condition of the gland, and this circumstance 
 made it difficult sometimes to estimate the size of a single acinus. 
 The cells within the acinus were many of them the same as the 
 round or oval nuclear bodies of the restinsj state, although more 
 numerous ; they occupied the cavity of the acinus generally. But 
 among the nuclear cells were a few of a totally different kind. 
 These are no other than the large yellow granular cells that have 
 been already described as characterising the later stages of involu- 
 tion. Fig. 12 is a drawing of two adjacent acini from the mamma 
 of a cat with foetuses 2^ inches long; it will be seen that one 
 acinus contains three of the large pigmented cells and the other 
 a single large cell. It is to be added, in explanation of the figure, 
 that the large cells have exactly the same yellow colour as the 
 
Cii. ir.] OF THE BREAST. 41 
 
 corresiDOuding cells of involution, and that some of thcni appear 
 to have two or even more nuclei at Fig. 12. 
 different parts of their periphery. It is -'^^^3: . /^^^~r^ I 
 necessary, before proceeding further with ^fc'OVv''"')' v^ "^ 
 the account of these earliest products of '^^^"<' •• 
 evolution, to repel an objection that may 
 be made. There is no danger of mis- 
 taking them for the last products of «,■•<, ♦,, 
 
 vtuviijQ ^liv^ •. ^ I A pair of acmi from the mamma 
 
 involution, the existence of which has pf \ caj; pregi.a.it with fa-tuses 2j 
 
 iii V vfiiAUj-v^u, v.i.i>^ ^^^.. Kjij^ v^ inches long. The acinus on the left 
 
 1»ppn nlrpidv nrnvpd Tt is tvilP indeed '^^"'^ contains three large yellow 
 
 ueen aucaoy pioveu. l\ ib uue tnueeu ^,^.j,g^ ^^^^ ^^^^^ ^^^ ^,^g ^.^i^^ ^^^_ 
 that traces of pigment remain in the tains one. Magnified soo diameters. 
 gland after the involution process has been completed ; but it is 
 either in the form of the entire pigmented cells which are still 
 detained outside the lobule in their passage from the gland, or it 
 is in the form of grains of free pigment scattered over the surface 
 of a lobule, or lying among the resting cells of an acinus. Except 
 under those circumstances the yellow pigment is not present in 
 the gland after involution is completed, and there can be no doubt 
 that the large yellow cells found within the acini in the early stage 
 of evolution are characteristic products of that stage. 
 
 In the three cases witli small foetuses, grouped together as 
 representing the first stage of evolution, the pigmented cells were 
 found in large numbers, indicating a very considerable cellular 
 productiveness of the acini, and there were certain instructive 
 differences as to the situations in which they were found. In the 
 most advanced of the three, the lobules were considerably ex- 
 panded, and the acini showed a distinct central space and a 
 regular arrangement of the cells round the wall ; there was in 
 fact an appreciable advance in evolution, and the condition came 
 very near that which will be described as characterising the next 
 definite stage of the process. Now, when the teats of this animal 
 were squeezed, during life, to find out whether the gland con- 
 tained milk, there issued from the teats not a fluid substance, but 
 a solid friable substance of a yellowish-brown colour in the shape 
 of cylinders or plugs, 'which broke off in short pieces as they 
 reached the orifice of the nipple*. On examining this substance 
 with the microscope, it was found to consist of rounded masses 
 or oblonof casts, in neither of which was the form of individual 
 ^ I am told that the same thing is of regular occurrence in cows that are in calf. 
 
42 
 
 THE PERIODICAL EVOLUTION 
 
 [Pt. I. 
 
 cells easily made out. On proceeding with the dissection and 
 exposing the mainmse in situ the ducts were seen radiating from 
 the periphery towards the nipples, as they are often seen where 
 the gland has not its full expansion; but in the present instance 
 they were obviously filled with a yellow material, the same as was 
 expressed from the nipple. The yellow contents were readily seen 
 in the stained sections of the prepared gland, and they were found 
 to consist of closely-packed masses of the large yellow cells. In 
 the same preparations the yellowish cells were seen to be still 
 within the acini at certain points; they were also found, as they 
 very commonly are, in the interlobular fibrillar tissue. But the 
 
 Fig. 13. 
 
 I'M'"' 
 
 
 From the mamma of a cat pregnant with fotuses ti inches long. 
 Obli(|ue section of a duct, showing collection of large yellow cells in its 
 submucosa, and the same bursting into its lumen. Magnified 150 
 diameters. 
 
 most remarkable position in which they were found was the loose 
 submucous tissue of large ducts. There they occurred closely 
 packed together in one or more rows, and in two or three places 
 where they were collected in the submucosa in largest numbers, 
 was observed the singular appearance of which Fig. 13 is a 
 drawing. A rounded cluster or knob of the characteristic yellow 
 cells, for the most part with a nucleus visible, is seen to have burst 
 through the epithelium into the lumen of the duct, and to be 
 
Cu. II.] 
 
 OF THE BREAST. 
 
 43 
 
 foUoAved up by several rows of the same cells in the submucosa, 
 converging towards the breach in the epithelial wall of the duct 
 through which the others had already passed. There is no doubt 
 that a very large number of the pigmented cells in this case left 
 the gland by way of the ducts, and that they entered the latter, 
 not directly from the acini, which were probably not in open com- 
 munication with the ducts at the time when the vellow cells were 
 
 Fig. 14. 
 
 From the mamma of a cat in an early sta^e of evolution. The left- 
 hand side of figure shows portions of two lobules, and the rest of the 
 lisjure shows the fat tissue outside the secreting structure crowded with 
 the large yellow cells. Magnified 90 diameters. 
 
 being formed, but by first escaping through the wall of the acinus, 
 then traversing the submucous tissue of the duct, and finally 
 breaking through into the latter at a convenient place. But 
 that is certainly not the only mode of their escape from the gland; 
 
44 THE PERIODICAL EVOLUTION [Pt. I. 
 
 it lias been observed in the preparations of this one case only, and 
 it probably belongs to a later period of the evolution when the per- 
 fectly free communication between the acini and the ducts is being 
 re-established. Fig. 14 is from another case in the cat, with foetuses 
 2^ inches long, and though it will have to be referred to again, it 
 is introduced here to show the very considerable numbers in which 
 the large pigmented cells are thrown off in evolution, and the 
 mode of their disposal. The left-hand side of the figure shows 
 portions of two lobules; with a higher power it may be seen that 
 some of the acini contain the large yellow cells in their interior, 
 and, indeed, the two acini of Fig. 12, which contain several yellow 
 cells, are drawn from another part of the same preparation. Be- 
 tween the two lobules the fibrillar tissue is occupied with a pro- 
 cession of those pigmented cells, along with a number of round and 
 elongated nuclear cells, which have no pigment around them. The 
 right of the figure shows an enormous collection of large nucleated 
 cells, which are all of the same pigmented kind. They have pene- 
 trated into the fat tissue about the gland in the manner that the 
 figure shows. The cells have their nuclei almost uniformly at one 
 end or on one side; several of the largest of them have two nuclei, 
 also excentric, and a few of them are distinctly vacuolated, the 
 vacuole containing a homogeneous fluid substance of the prevailing 
 yellow colour. The case will not be further described until the 
 general question of the disposal of cellular waste by the lymphatic 
 channels is considered in the next chapter. 
 
 Proceeding to the next well-marked stage of evolution, a 
 certain definite appearance of the gland is found to be common 
 to the cases where the foetuses were 3 to 4 inches long, and. an 
 approximation to it is found in some jDarts of the gland in the most 
 advanced of the group of three cases representing the first stage. 
 Fig. 15 represents this characteristic appearance. The acini are 
 about two-thirds of their full size, and their arrangement in 
 clusters within the lobule can be seen more clearly than in any 
 other condition of the gland. The cells within them are still of 
 the nuclear kind, that is to say, they have no broad fringe of 
 protoplasm round them; and although there appears to be a de- 
 finite short interval between each cell, those intervals cannot be 
 .said to be occupied by the cell-substance. They are, however, dis- 
 tinctly larger and more granular than the cells of the resting state, 
 
Ch. 11] 
 
 OF THE BREAST. 
 
 45 
 
 Fig. 15. 
 
 
 
 From the mamma of a eat preg- 
 nant with fii'tuses Si iiichi'S loiip. 
 Someof theacini coiitaiTia grniimlHr 
 material. MagiiiliedSOO diameters. 
 
 and than many of the same class of nuclear cells iu the stage that 
 has just been described. They form a 
 complete covering to the floor of the 
 acinus, being arranged, as has been said, 
 in regular order with definite intervals 
 between them. It can readily be con- 
 ceived that this orderly arrangement 
 depends on the re-establishment of the 
 regular meshes of the basement mem- 
 brane, although the latter are not easily 
 made out. A very regular polygonal 
 meshwork of transparent fibres of uni- 
 form thickness has been seen in the 
 very small acini of one of the cases in 
 the tumour series, where the gland was 
 in a condition resembling that of an 
 early stage of evolution. Besides the 
 regular order of the cells in the floor of 
 the acinus, those forming the circuit 
 round it, as seen in the profile view, are 
 
 also placed in a regular line with short spaces between them. 
 Another distinguishing feature of this stage is the presence in the 
 cavities of the acini of a peculiar granular material, the coagu- 
 lated condition of a fluid. In many acini it takes on the colour 
 of the staining agent to a certain extent, in others it retains the 
 character of highly-refracting granules, while in others it is found 
 to be mixed with or to pass gradually into a fluid substance which 
 is often vesiculated. 
 
 In the glands now under consideration, corresponding to 3 to 4 
 inches length of foetus, the evolution has advanced to a point when 
 there is no further trace of the large pigmented cells, whether in 
 the acini or in the interlobular fibrillar tissue. In the former 
 group of three cases, showing the early evolution changes, there 
 was one that came nearest to the later stage of evolution, now 
 being described; and in that case the pigmented cells were already 
 collected in large numbers in the ducts, and were seldom met with 
 at or near the place of their origin. The season of their produc- 
 tion from the acini had already passed. But in the two other 
 cases of the same group, which were known from other circum- 
 
46 THE PERIODICAL EVOLUTION [Pt. I. 
 
 stances to be at an earlier stage, the pigmented cells within the 
 acini were comparatively numerous. But although the later stage 
 of evolution is not characterised by the presence of large pigmented 
 cells in the tissues of the gland, it is characterised by the presence 
 of no less important cells of another kind that are also to be seen 
 collected in heaps or in transit along the interfascicular spaces. 
 The absence of pigment in these cells makes it a more difficult 
 task to prove their identity : that is, to prove where they come 
 from and where they go to. There is, however, no room for doubt- 
 ing the existence of a general class of waste-cells of the secretion ; 
 the woodcut on page 48 shows an unmistakeable collection of 
 them. If all the cells that fell to be described under the category 
 of waste-cells were of the pigmented kind, the doctrine of a cellular 
 bye-product of secretion might be established by a simple demon- 
 stration. So far as the pigmented cells go, they" are evidence of a 
 crude or cellular secretion produced at a time when the function 
 has all but come to rest, and again when it has just begun to 
 revive. That there should be corresponding crude products of the 
 secretion at other stages of the subsiding or reviving function is in 
 itself highly probable. The difficulty of actually proving their ex- 
 istence lies in the circumstance that they have no distinctive mark 
 such as the pigment of the more immature cells, and that they are 
 in size and form much the same as connective-tissue cells, fixed or 
 amoeboid, or as emigrated colourless corpuscles of the blood. The 
 last-named need not be seriously considered ; they are not all 
 white corpuscles that have the appearance of white corpuscles. 
 But the connective-tissue cells are a real source of ambiguity, 
 and that ambiguity is well illustrated in the subjoined woodcut 
 (Fig. 16). It cannot be said that the appearance therein drawn is 
 of so convincing a kind as many other parts of the glands might 
 have furnished. But it represents, at least, a common appearance 
 in the interlobular tissue of breasts at the particular stage of evolu- 
 tion wdiich, in the cat, corresponds to foetuses about three inches long. 
 The figure shows the interlobular tissue, and to a certain 
 extent also the interacinous tissue, to be crowded wdth oblong 
 or spindle-shaped cells of granular substance and of considerable 
 bulk or plumpness. They appear to lie loosely in the interfibrillar 
 spaces. If they are to be compared to the connective-tissue 
 cells of a healthy part, the comparison can only be with embryonic 
 
Ch. II.] OF THE BREAST. 47 
 
 counective-tissue cells ; and in the mamma . of a very young- 
 animal, even perhaps of a primipara, there would be a strong 
 presumption that the cells belonged to the connective-tissue 
 stroma. The animals in question are all marked in my notes 
 as being full-grown, and I have no further information about 
 them. 
 
 Fig. 16. 
 
 From the mamma of a cat pregnant with foetnsps 3 inchf s long. At 
 the upper border is represented the outermost row of acini of a lobule; 
 on the left, below, a small lobule; at the two sides, curves corre- 
 sponding to the boundaries of two other lobules ; in the centre, a duct 
 with a group of acini round it; the interlobular tissue cnnvded with 
 oblong cells. Magnified 160 diameters. 
 
 There is, however, a considerable body of evidence to support 
 the view that the oblong and other cells in the interfibrillar 
 spaces belong to the same class of waste-cells of the secretion 
 as the unmistakeable pigmented cells already described. The 
 strongest evidence is perhaps the fact that they are very commonly 
 found mixed with the pigmented cells, and that the latter often 
 assume the same oblong or fusiform shape. The left-hand side 
 of Fig. 14 contains both kinds of cells crowded toofether. 
 
 Again, they are found in heaps at various points in the inter- 
 lobular tissue, and in such accumulations it is hardly possible 
 to mistake them for the fixed elements of the stroma. The 
 heaps are circumscribed in extent, and the cells composing them 
 are arranged in rows corresponding to spaces between the fibres 
 of the stroma. The forms of the cells, also, are much less am- 
 
48 THE PERIODICAL EVOLUTTOX [Pt. I. 
 
 biguous tliau those drawn in Fig. IG; they are rounder and 
 sometimes cubical, and may be compared to the class of " paren- 
 chyma-cells " which have been described as part of the inter- 
 tubular or inter acinous structure of other glandular organs\ 
 
 Another point in the evidence is that when the breast is 
 examined at a later period of evolution, not indeed in the same 
 animal but in animals of the same size and otherwise resembling, 
 the stroma is found to be comparatively free of all such cells. 
 But all of those more or less dubious proofs are superseded by 
 the striking evidence derived from pathology. In the patho- 
 logical processes, to be afterwards described, which have their 
 physiological type in the normal process now spoken of, there is 
 the clearest proof of the epithelial origin of cells in the stroma 
 other than the pigmented cells. Not only oblong or fusiform 
 cells, but large spherical and even cubical cells are found in rows 
 and heaps outside the acini of the gland, and having a perfect 
 likeness in all points to as many varieties of cells within the 
 acini. To invoke the evidence of pathology at this stage is by 
 no means to argue in a circle ; there are good reasons why the 
 morbid physiological processes should in some points more 
 clearly reveal the law of the function than the healthy processes. 
 The pathology is not borrowed from the physiology, but it hangs 
 together with it. In like manner the phenomena of the sub- 
 siding process accord with those of the reviving process ; and 
 the facts relating to the disposal of the cellular waste of the 
 secretion by means of the lymphatic glands complete the con- 
 currence of testimony. Certain parts of the doctrine of waste 
 cells of the secretion may be histologically weak, but they are 
 
 1 I have lately met with the two following references to cells in the stroma, pro- 
 bably of the same nature as those described in the text ; the authors, however, do 
 not appear to have assigned any clear physiological significance to them, but have 
 described them, after the "objective" fashion, as noteworthy points in the histology 
 of the breast, and without reference to its periodical processes. Winkler [Jahres- 
 bericht d. Gesel. f. Naturwis. Dresden, 1874), describes " leucocytes " in the walls of 
 the alveoli, and suggests that they play a part in the formation of the milk. Von 
 Brunn (GottingerNachrichten, No. 19, 1874) found in the interstitial connective tissue 
 of the breast in man and in the rabbit, and in that of the submaxillary gland in the 
 ox, cells and groups of cells of the same character as those described by Hofmeister 
 and others in the testicle {Parenchym-zellcn). They were rarely found in the 
 mamma of the virgin. Von Brunn would also connect them in some way with 
 the secreting cells. 
 
Cii. IL] 
 
 OF THE BREAST. 
 
 49 
 
 entitled to the support that arises out of a generalisation on 
 the whole well established. 
 
 We come now to the later stage of evolution, immediately 
 preceding the actual formation of milk. Fig. 17 is a drawing 
 which might have been made from any part of the mammae of 
 cats that were pregnant with foetuses almost full grown. The 
 same appearance might also have been drawn from many parts 
 of the udder of a ewe that was pregnant with a foetus about a 
 foot long. At the stage of evolution, the description of which 
 we have now reached, the glands are voluminous, and of the 
 almost uniform parenchymatous consistence of the fully expanded 
 
 Fig. 17. 
 
 ./o-M^M-i^^.- (kill. ' '''' f^^^ 
 
 
 From the mamma of a rat prpgiiant with fuetuses almost full trrown. 
 Vacuolation of the epithelium; two acini on the left coniaininj< mucus. 
 Magnified 300 diameters. 
 
 state. In the case of the ewe just mentioned, a clear brownish 
 fluid could be expressed from the teats. On microscopic examina- 
 tion, the lobules are fully expanded, and the interlobular tissue 
 forms quite thin septa. The entire field of the microscope is 
 occupied by the closely-packed acini, which are of the full size. 
 The appearance that is characteristic of this stage is given in 
 Fig. 17. Many acini are filled with a mucous fluid, colouring 
 with the staining agent, and often penetrated with a large number 
 of vesicles. The floor of the acinus, and the circumference of 
 it when seen in profile, are occupied by a complete set of cells 
 in regular order, all of which are vacuolated, and most of them 
 c. c. 4 
 
50 THE PERIODICAL EVOLUTION [Pt. I. 
 
 in exactly the same manner. If the figure is referred to, it will 
 be seen that the peripheral mass of the vacuolated cell is almost 
 always large ; it has a somewhat oblong or reniform shape, and 
 shows a nucleolus. Looking at the floor of an acinus, an im- 
 pression is got which it has been found difficult to convey in the 
 drawing. The excentric nuclear bodies all look not so much as if 
 they were situated on the periphery of a round vacuole, as that 
 they lay each on one side of the polygonal meshes of the basement 
 membrane. This simply re-opens the former question of the 
 relation between the polyhedric form or outline of the cell, and 
 the regular polygonal space in the basement membrane in wliich 
 it is set. But if the profile view of the epithelium is looked 
 at, it becomes clear that we have to deal with the same vacuo- 
 lation as elsewhere ; and the peculiar appearance of the floor of 
 the acinus must be owing to the prominence of the fibres of 
 the basement membrane, side by side with the delicate vacuolar 
 outlines. Tlie mucous fluid in the cavities of the acini is ob- 
 viously the vacuolar product of the epithelium, and it is in- 
 teresting to note that the period at which mucus was found 
 in the acini in the course of involution, corresponded to exactly 
 the same degree of vacuolation of the cells. It is the period 
 of involution that is represented in Figs. 5, 6 and 9, the two 
 former being from a cat nine days after the end of lactation, and 
 the latter from a sheep that was killed three weeks after par- 
 turition, without having been milked. 
 
 It is a somev.'hat remarkable circumstance that the case of the 
 ewe with a foetus about a foot long, and therefore far advanced in 
 pregnancy, shows in some jDlaces exactly the same enormous collec- 
 tion of small round cells about the acini as in the case of the ewe 
 in the involution process from which Fig. 9 is taken. In both 
 cases the small round cells are, in isolated portions of the gland, 
 so numerous as to completely obscure the outlines of the acini. 
 There is, no doubt, something unusual in the circumstances of each 
 case ; but they none the less illustrate the physiological law of 
 their respective processes, and the remarkable parallelism between 
 the same\ In the cases of advanced evolution in the cat, as repre- 
 
 1 In the one case, the ewe was kept at the Brown Institution, London, among 
 others that were used for the experimental study of sheep-pox, hut it was not itself 
 inoculated. It died from some unknown cause (probahly sheep-pox acquired iu the 
 
Ch. II.] OF THE BREAST. 51 
 
 sented by Fig. 17, there is little or no appearance of cellular waste 
 products outside the acini, and within the acini it is only the 
 fluid product that is to be seen. Whether the cellular masses that 
 occupy the periphery of tlie vacuoles in this stage are ever dis- 
 engaged and thrown off as cellular waste, or whether the cell as 
 now seen is only partially vacuolated but destined to be entirely 
 so, is not easy to decide. It is certain that there are found side 
 by side with those signet-ring like cells, a few that are nothing 
 but delicate ring-like forms, without any cellular enlargement on 
 their peripher3\ Again, if there is any cellular waste at this stage 
 of evolution, it would be discharged by the ducts, along with the 
 mucous or fluid products of the epithelium. 
 
 There is but a single step from the period of evolution just 
 described to the jDerfect state of the epithelial cell which is found 
 a day or two before parturition. The appearance of a mosaic of 
 polyhedric cells with a central nucleus may be seen in occasional 
 acini of the previous stage, but the protoplasm of the cell is 
 perfectly hyaline and not at all granular ; the nucleus seems 
 rather to lie in the centre of an empty polygonal space. About 
 the time of parturition the epithelium has acquired its granular 
 appearance, and shortly after the birth of the offspring the 
 milk beg'ins to be formed in great abundance. The milk of 
 the first few days contains a large number of the well-known 
 colostrum cells, which are the last of the long series of imperfect 
 secretory products that have been thrown off during the evolution 
 process. The appearance of well-defined colostrum cells is very 
 much the same as that of the cells in Fig. 8, which are found 
 within the acini in the earliest stage of the involution of the 
 gland. 
 
 The method followed in the account of the involution process, 
 of giving a description of the gland at various well-marked stages 
 of its progress, without any commentary as to the physiological 
 significance of the same, has been mainly followed also in the 
 account of the evolution ; and if it be added, that the well-marked 
 stages that have been chosen for description are of necessity some- 
 ordinary way of contagion) ; the lungs were full of a wMtish material, and the uterus 
 contained two nearly full-grown lambs that had been dead some days. The other 
 ewe was inoculated with sheep-pox on Dec. 8th ; it recovered completely, and wa; 
 killed on March 13th, having given birth on Feb. 2Gth to a dead lamb full grown. 
 
 4—2 
 
52 THE rEUTODICAL EVOLUTIOX [Pt. T. 
 
 what conventionally divided from each other, the preceding account 
 of evolution may be taken as a fairly complete statement of the 
 facts of the case. It now remains, as in the chapter on involution, 
 to consider, by way of summary, what is the physiological law of 
 the gland's activity during the restoration of its structure. 
 
 It has been explained, in the first chapter, that it took several 
 weeks for the secretor}' activity of the mamma to run down or to 
 subside, that the gradual subsidence was accompanied with the 
 production from the epithelium of a good deal of cellular waste, 
 and that the conclusion of the process of involution left the gland 
 with all the parts of its framework entire, but with the acini 
 greatly contracted and containing only a heap of nuclear cells. 
 The renewal of the gland's activity begins, roughly speaking, with 
 the commencement of the next pregnane}^ and it brings with it, 
 step by step, the restoration of the full epithelial structure which 
 had been lost during the previous subsidence. This rehabilitation 
 is in most animals a very slow process ; one may say that it is 
 unnecessarily slow, in respect of the object to be attained. But, 
 as a matter of fact, it is not completed till the term of gestation 
 is at an end. The evolution of the gland is, as it were, distributed 
 over the entire period of pregnancy. Starting then with the 
 resting state of the gland, we are met with the question. How do 
 the nuclear cells that are left occupying the acini at the end of 
 involution assume the character of nucleated polyhedric epithelium, 
 and according to what law of cellular activity does this renewal 
 take place ? Now, as was contended mutatis mutandis in the 
 analogous case of involution, it is not to be supposed that the 
 particular set of nuclear cells with which an acinus is furnished 
 when the unfolding process begins are the same that, after various 
 transformations, spread over a very long period, become the poly- 
 hedric epithelium of the lactation stage. On the contrary, there 
 is no more certain fact in the study of the periodical mammary 
 evolution than the Lirge production from the acini of various kinds of 
 waste-cells at various stages of the process ; the figure No. 14 is well 
 adapted to show a large collection of cells that have been thrown 
 off from the gland. It is in the successive renewals of the cells 
 within the acini, which this fact involves, that we find the succes- 
 sive steps towards perfect epithelium. Each new generation of 
 cells is more like the polyhedric nucleated epithelium than the 
 
Ch. 11. ] OF THE BRKAIST. 53 
 
 preceding. One knows no process of simple multiplication of cells 
 by which the nuclear elements of the resting state could produce 
 polyhedric epithelium, even if the process were continued through 
 many generations. If they multiplied, as no doubt they do, they 
 would produce nuclear cells like themselves ; and if we were 
 limited, in our explanation, to a mere process of proliferation or a 
 formative process, it would be impossible to understand how those 
 nuclear cells ever came to have such descendants as the perfect 
 epithelial cells. But we have already seen that the activity of the 
 rbamma is always double ; there is both a formative action and a 
 functional action, and those two, in health, always go hand in 
 hand. Therefore, when the re-awakening forces of the breast are 
 directed to the cells within the acini, the latter not only multiply, 
 but they are at the same time transformed in the direction of the 
 secretion. It is the transforming force and the increasing inten- 
 sity of the same that is the factor wanting to explain how the 
 nuclear cells at one end of the scale are the progenitors of elements 
 so unlike themselves as the perfect epithelial cells at the other 
 end. The unfolding of the apparatus of secretion is necessary to 
 the revival of the function, and, on the other hand, the return of 
 the cells to their perfect form could not take place were it not for 
 the increasing intensity of the secretory force. This is the dualism 
 of structure and function, the highest correlation in our knowledge 
 of the organism, the double name of "two things mutually ex- 
 plaining and equally real." The dualism of structure and function 
 is therefore not to be resolved into terms more ultimate, and it is 
 in that sense inexplicable. But it is at least an illustration of the 
 union between structure and function to point out that the one 
 varies as the other varies ; that a low degree of function resides in 
 what may be called a crude shape of cell, that a high degree of 
 function resides in what may be called a perfect shape of cell, 
 and that the intensifying of the function and the perfecting of 
 the cell go hand in hand. 
 
 The restoration of the resting cell of the acinus to its poly- 
 hedric epithelial shape does not take place in one generation of 
 cells, but in the course of several generations ; the intermediate 
 generations of cells are, from the other point of view, the waste 
 products of the crude secretion. The earliest change noticeable 
 in the unfolding of the breast is the presence within the acini of a 
 
54 THE PERIODICAL EVOLUTION [Pt. 1. 
 
 limited number of large yellow granular cells. Those cells were, 
 in the former chapter, found to be the last products of the declin- 
 ing secretory force in the upfolding process of the gland, they are 
 the first products of its re-awakening in evolution ; and it will be 
 shown in the 2mthological part of the work, that the same trans- 
 formation of the epithelium is the characteristic effect of any 
 feeble stimulus whatever, applied to a gland in the resting state. 
 
 The first step then in the periodical evolution of the gland 
 is the transformation of nuclear cells of the resting state into 
 large granular cells containing yellow pigment. But not all 
 the cells of an acinus, or at any rate not all of them at once, 
 assume the enlarged and pigmented character. Just as we saw 
 that some at least of the resting cells of the acini were derived 
 from the large yellow granular cells of involution by the latter 
 losing their pigmented granular substance and leaving their 
 nuclei behind them, so it now appears that certain of the same 
 nuclear resting cells recommence the cycle of change by acquiring 
 a granular and pigmented investment. In some such way, at 
 least, must one construe the re-appearance within the acini of 
 the large pigmented cell in the early stage of evolution. The 
 change from a nuclear to a large granular condition is observed 
 elsewhere in the adult body; Virchow in particular has remarked 
 on the reality of this change. Combating the mistake which, 
 as he says, is commonly made of treating white-blood corpuscles, 
 lymph-corpuscles, and the cells of the follicular tissue of lymphatic 
 glands as if there were no morphological differences between 
 them, and as if they were identical, he points out that there is 
 a distinct act of transformation before one of these can become 
 the other, and that this transformation is no less real than the 
 passage of a cell of the rete mucosum to become an epidermic 
 cell. For example, the commonest kind of cell in the parenchyma 
 or follicular tissue of a lymphatic gland is a "naked" nucleus, 
 without an obvious cell-body. The cells of the lymph, on the 
 other hand, have a more voluminous cell-body, which is also so 
 much more dense that the nuclei become less distinct. Still more 
 is this the case with the colourless blood-corpuscles, whose dense 
 and exceedingly granular bodies quite conceal their nuclei, so 
 that the latter are made visible only by reagents \ 
 
 1 Cellular-Patholofjie, 4tb cditiou, pp. 210—212. 
 
Cu. II. J OF THE BREAST. 55 
 
 That is no doubt an analogous change under somewhat 
 different circumstances. In the embryonic development there 
 is hardly a tissue of the body but would show similar changes 
 in its cells, as they pass from their indifferent nuclear form to 
 their characteristic mature form. In the present case the increase 
 in size is both very great and apparently very sudden, and the 
 swollen cell-substance is occupied by yellow pigment. I have 
 already mentioned that only one or two cells in an acinus at one 
 time appear to be subject to the reviving secretory force, and 
 that the large yellow cell often includes more than one nucleus. 
 
 The pigmented cells are thrown off from the acini in large 
 numbers ; in the case from which Fig. 14 has been drawn, a 
 large collection of them had been accidentally formed in the 
 fat-tissue just outside the secreting structure. It is probable that 
 they escape from the acini by passing through their walls, the 
 still upfolded acini being not yet in free communication with 
 the lumen of the ducts \ If that collection be examined with a 
 higher power, it would be found that some of the cells have two 
 nuclei, that the nuclei are all excentric and therefore clearly 
 to be seen, and that a cell here and there has undergone the 
 typical vacuolation, the fluid in its vacuole being of the same 
 yellow colour as the granular cell-substance of the others. The 
 pigmentation does not continue into the later stages of evolution, 
 just as it did not commence till towards the end of involution. 
 As the functional stimulus gets gradually stronger, the cells that 
 are produced within the acini, whatever more epithelial form 
 they may have attained to for a brief space on the acinus wall, 
 are thrown off as somewhat large nuclear bodies of a granular 
 appearance, which often acquire an elongated or spindle form 
 in their passage through the interfascicular spaces. The difference 
 between them and the pigmented class of products, is that they 
 have undergone a certain amount of functional transformation 
 within the acini, as is shown by the presence of a sometimes 
 granular and sometimes fluid material in the acinous cavities ; 
 
 1 Astley Cooper (Anatomy of tlie Breast), usiug the now obsolete names for 
 glandular structures, says : " In the fulness of lactation, those leaves [lobules] are 
 full of cells [acini], which can be readily injected and demonstrated; but at other 
 periods they do not admit of being filled, and a most minute injection may then 
 be made of the lactiferous tubes, yet no cells [acini] appear." 
 
50 THE PER IODIC A L EVOLUTION OF THE BEE A ST. [Ft. I. 
 
 whereas the large granular pigmented cells are tlirown off un- 
 changed. It may be recalled to mind once more that the pig- 
 mented cells do show a tendency to become vacuolated, and 
 otherwise to lose their cell-substance after they have left their 
 l^roper functional habitat. 
 
 Coming to the last stages of evolution, we again reach familiar 
 ground. In the glands corresponding to almost the full size of 
 the foetus, there is perfectly typical vacuolation of the epithelium 
 going on, with the formation of mucus as the fluid product ; 
 and that vacuolation takes place in cells that have now recovered 
 the polyhedric form of epithelium. By this time the secretory 
 excitation of the gland has increased so much in intensity that 
 the functional transformation is only slightly different from that 
 of the perfect state. Those slight differences are however in- 
 structive. They enable us to interpret aright the nature of 
 the secretory process when the function is at its height, and 
 when the transformation of the epithelium has no morphological 
 features; and they equally help to clear up the obscurities of 
 the process in its earliest stage. The parallelism with the in- 
 volution process is complete. We again find ourselves to be 
 following a progressive series of vacuolation changes, with a 
 correlation between the fluid and cellular elements of the vacuola- 
 tion. At one end of the series, everything is vacuole, and there 
 is no cellular production ; at the other end of the series, everything 
 is solid or cellular, and there is no vacuole. The glandular 
 activity in its re-establishment, no less than in its subsidence, 
 follows the law of endogenous cell-formation, and that is also 
 the law of its action when it is in its full career. Again, the 
 periodic evolution is accompanied by much cellular waste ; and 
 the cellular waste of the evolution is more than that of the in- 
 volution, in proportion as the former process is spread over a 
 longer period. 
 
CHAPTER III. 
 
 THE LYMPHATIC GLANDS OF THE BREAST IN CONNEXION 
 WITH THE DISPOSAL OF ITS CELLULAR WASTE. 
 
 The main physiological conclusions of the two preceding 
 chapters may be said to be the performance of the mammary 
 function according to the law of endogenous cell-formation, and 
 the production, during involution and evolution, of a large amount 
 of cellular waste, arising from the operation of that law in the 
 subsiding or immature periods of the function. Of those two 
 conclusions, one may be said to be more theoretical and the other 
 more practical ; if the theoretical conclusion be held in doubt, 
 the facts relating to the waste-cells may be taken without theory. 
 The present chapter will be devoted to a consideration of the 
 further progress and ultimate disposal of the waste-cells in the 
 lymphatic glands. 
 
 It is necessary, at the outset, to recapitulate the various 
 forms of cells thrown off at different periods. They may be 
 classed in three groups, subject always to the reservation that 
 there is a gradual transition between them. The first kind of 
 waste products are the familiar colostrum cells, which approach 
 most nearly, whether at the beginning of involution or the end 
 of evolution, to the perfect secreting cell. The second kind 
 are the more or less nuclear bodies of somewhat various shapes, 
 which are the survivals of well-marked vacuolation of the epi- 
 thelium and are usually associated with a mucous fluid ; they 
 characterise the middle period of both processes. The third 
 kind are the very remarkable large pigmented cells, which are 
 the latest products in involution and the earliest in evolution. 
 
58 THE LYMPHATIC GLANDS [Ft. I. 
 
 The disposal of the colostrum cells at the end of evolution (be- 
 ginning of lactation) is a matter of common observation. They 
 are got rid of by the ordinary outlets of the secretion, viz. the 
 ducts, and they are found mixed with tlie first milk in large 
 numbers. The corresponding imperfect products at the close of 
 lactation are no doubt disposed of in the same way, although 
 their discharge is not accelerated, as in the former case, by the 
 strong flow of the rising secretion. The ducts are, however, found 
 for several days after lactation, and even as late as the second 
 week, to be filled with a milky substance, and in some cases 
 a fluid escapes from the breasts for a considerable time after 
 weaning. The later or non-vesiculated kinds of waste-cells are 
 got rid of usually in a different way. Preparations have already 
 been described and figured which show them to be collected 
 about the acini and in the fibrillar tissue. The acini, at the 
 periods when those cells are produced, do not appear to com- 
 municate freely with the ducts; the cells therefore find their 
 way into the lymphatic channels of the gland, whence they are 
 in health ultimately drained off. All the waste-cells of the 
 cruder sort, though they differ among themselves in size and in 
 other respects, may be classed together for the present purpose. 
 But, as might be surmised, it is the pigmented cells that are 
 again the least doubtful sources of evidence, and they will be 
 made to sustain the proof of the statements that apply to both 
 kinds alike. They are perhaps not the most abundant of the 
 waste-cells, but their unmistakeable yellow colour makes them 
 by far the most valuable for demonstration, and enables the whole 
 question of the disposal of the cellular waste to be treated much 
 more objectively and briefly than it could otherwise be. 
 
 Now, as regards the large yellow granular cells, they are 
 most commonly found outside the acini in the position given in 
 Fig. 10. Having once entered the lymphatic spaces, they for 
 the most part continue their progress in the same. But a case 
 has already been described and figured which constitutes an ex- 
 ception to the general rule. In it the yellow cells are also found 
 to have escaped through the walls of the acini, but they travel 
 along the submucous tissue of the ducts, and at convenient points 
 break through into the latter, which were found throughout the 
 gland to be filled with them. Apart from this case, there is the 
 
Ch. III.] OF THE BREAST. 59 
 
 clearest evidence, collected from a large number of preparations, 
 that the pigmented cells are conveyed directly from the mamma 
 to lymphatic glands situated, in the animals enumerated, at each 
 end of the mammary chain. Those at the upper end are the 
 axillary glands ; those at the other end are the glands more 
 particularly referred to in the sequel, and they require a few 
 words of description. 
 
 It has been already mentioned that the inguinal portions 
 of the mamma rest on cushions of fat, which are sometimes of 
 great thickness. Thus, in two cases of advanced involution, in 
 the dog and the cat, the lower end of the mamma,ry chain, forming 
 a thin layer of tough substance, rested on a cushion of ftit about 
 three-fourths of an inch thick. In each of the cases there were 
 found in the centre of the fat one or two quite small masses of 
 tissue, which proved on examination to be lymphatic glands. In 
 most of the other cases the lymphatic glands have been found 
 of larger size, and placed close to the under surface of the mamma 
 and towards its outer border, where also the fat extends. Some- 
 times a chain of them is found extending beneath the mamma 
 for two inches or more forwards along the abdominal wall, and 
 these also are embedded in a certain quantity of fat. They 
 are for the most part long, narrow, and flattened cylindrical 
 bodies, with a hilus extending alonsr the whole leng-th of one 
 edge. The appearance of a hilus is produced by the apposition 
 of two rounded borders, as if the gland had originally been a 
 flat disc-shaped body like the spleen, and had become folded upon 
 itself. Along the hilus border there are always found the cut 
 ends of an artery and a vein of considerable size. 
 
 From their position, these lymphatic glands seemed to belong 
 to the mamma in some especial way. They were found to vary 
 much in size in the different cases, but it has not been possible 
 to make out that they vary in any exact measure with the 
 periodical changes in the mamma itself. However, their relation 
 to the mamma is put beyond question even by a superficial 
 inspection. One particular case out of a large number is sin- 
 gularly adapted for demonstration, and it will be briefly described. 
 It occurred in a bitch that had a large tumour occupying a certain 
 limited portion of the mammary chain on one side. The animal 
 was killed, owing to the incurable nature of the disease (secondary 
 
60 THE LYMPHATIC GLANDS [Pt. I. 
 
 tumours in the liver and lung), and the whole chain of glands 
 on both sides was dissected off. The new growth was perfectly 
 circumscribed ; and the entire mamma on one side and the 
 greater portion of it on the other were found to have the usual 
 tough fibrous appearance of the gland at or about the resting 
 state, although with an unusual amount of yellowish-brown 
 pigmentation. The attention is specially directed to the inguinal 
 portion of the gland on one side. It had the appearance, as has 
 been said, of a tolerably normal mamma at or near the resting 
 state ; it was about a line in thickness and lay upon a cushion 
 of fat half an inch thick. In the midst of this fat and at a 
 distance of about one-quarter of an inch or more from the 
 under surface of the gland, thei'e were found two quite small 
 parenchymatous bodies of a uniform yellowish-brown colour, 
 which proved on microscopic examination to be lymphatic glands. 
 One of them appeared as the cross section of a slightly curved 
 disc-shaped mass, and close to its concave surface there appeared 
 the cut ends of two large vessels. The other was a somewhat 
 pear-shaped body placed in the fat in an oblique position. 
 
 The microscopic examination, as has been said, showed them 
 to be lymphatic glands. But they did not contain the relative 
 proportions of the various tissues that are ordinarily met with 
 in lymphatic glands. The follicular tissue corresponded chiefly 
 to the thick end of the pear-shaped mass and to the convex 
 surface of the crescentic ; the cells of that tissue were of the 
 ordinary lymphoid kind. The surface of the follicles Avas very 
 commonly strewn with grains of brown pigment ; otherwise the 
 follicular tissue was not pigmented. Probably the greater part 
 of the glands consisted of the wide lymph-sinuses and the dense 
 fibrillar tissue of the hilus, through which the former penetrated. 
 The large lymph-sinuses were everywhere filled with the yellow 
 oranular cells that have been described and figured in the first 
 and second chapters. Fig. 20, on a subsequent page, is a drawing 
 of a portion of one of these preparations; the enormous pre- 
 ponderance of the stroma of the hilus and of the lymph-sinuses, 
 and the quite rudimentary condition of the follicular tissue, will 
 be at once apparent, but the further description of the preparation 
 is reserved till later. The yellow cells are represented in the 
 woodcut simply as granular. 
 
Cii. Ill] OF THE B UK AST. Gl 
 
 It has been already mentioned that the mamma of this case 
 was to the naked eye more pigmented than usual ; and on 
 microscopic examination many yellow or brown granular cells 
 were found in and around the acini, which were for the most 
 part only a little larger than in the resting state. A series of 
 sections were made which included in one piece the mamma, 
 the fat beneath it, and the lymphatic gland in the midst of the 
 latter; and it was possible, in one and the same preparation, 
 to identify in the clearest manner the pigmented cells of the 
 mammary acini with those in the lymph sinuses of the lymphatic 
 gland. Certain tracts of tissue were found running through 
 the fat between the mamma and the lymphatic gland, which 
 contained, no doubt, the collapsed lymphatic vessels. None of 
 the pigmented cells were found at any point between the mamma 
 and the lymphatic gland, and that circumstance will now be 
 in a manner explained by reference to a preparation from another 
 case. It is a case that has been already referred to, and from 
 which the figure on page 43 has been drawn. It represents 
 the early stage of evolution in the mamma of the cat, the foetuses 
 having been '2.^ inches long. The acini and surrounding tissue 
 contained numerous large pigmented cells, and so also did the 
 lymphatic glands of this case, a drawing from which latter is 
 introduced afterwards (Fig. 19). But the remarkable feature 
 of the case is the collection of the same yellow cells in the fat 
 tissue outside the mamma. It seemed at first sight as if this 
 were a case of the pigmented cells 'on their way through the 
 submammary fat to the lymphatic gland. But a second examina- 
 tion showed that such was not the case. In the inguinal mamma 
 of this case, as in one or two others, there was, in addition to the 
 fat beneath the mamma, a small quantity of fat on its upper 
 surface. Fat on the upper surface is unusual in animals that 
 have the extended chain of glands ; for the mamma, or its loose 
 fibrous investment, nearly always adheres closely to the skin 
 above. Now, on a second examination of the preparations, it 
 proved to be the external layer of fat, and not the cushion beneath 
 the gland, that contained the large collection of yellow cells. 
 It is not at all a fanciful account of the matter to say that the 
 pigmented waste-cells had made their escape from the secreting 
 acini on the wu-ong side of the organ ; they had travelled along 
 
02 THE LYMPHATIC (; LANDS [Ft. I. 
 
 channels that were no doubt open to them, but which led into 
 a layer of fat that had no further outlet. The unusual layer 
 of fat on the upper surface of the gland acted as a trap to a 
 certain number of waste-cells; if they had been conveyed like 
 the others into the fat ou the lower surface, they would have 
 been on the way to their proper reservoir, the lymphatic gland, 
 'i'lie detention of waste-cells in the fat in this case is of the 
 exceptional kind that proves the rule ; if there had been no 
 obstacle to their transit they would have passed singly and 
 (piickly through the lymphatic channels, so that the preparations 
 of the animal killed at a particular moment would probably 
 not have shown a single cell in its passage. The same difficulty 
 of finding cells in some particular act of their life constantly 
 occurs in the histology of the dead tissues. In histological in- 
 vestigation mere traces or indications of a cellular process must 
 often suffice as the basis of a general conclusion for which they 
 seem inadequate. The reasonableness of the conclusion depends 
 very much on the support that it receives from associated cir- 
 cumstances. In the present case the evidence amounts as nearly 
 as possible to demonstration ; the granular pigmented cells, in 
 one and the same preparation, are found within the secreting 
 acini, in the connective-tissue spaces outside the secreting 
 structure, and in the lymph-spaces of the subjacent lymphatic 
 gland. In some cases there has been observed also a collection 
 of cells in the tissue just outside the convex surface of the 
 lymphatic gland ; that is to sa}^, at the points where the in- 
 carrying lymphatic vessels enter. But in those cases the cells 
 were not of the pigmented kind, and their occurrence might 
 be explained in some other way. 
 
 So far as the facts have been already stated, the Ijauphatic 
 glands under the mamma are to be viewed as special appendages 
 of the secreting organ ; whatever other function they may have, 
 they are obviously there to receive the cellular waste products 
 of the secreting gland to which they are attached. In the case 
 described, where the lymphatic gland consisted to a great extent 
 of lymph-sinuses filled with the pigmented cells from the mamma, 
 it was impossible to resist the notion that the collection of the 
 cellular waste of the secretion and the disposal of the same was 
 its raison d'etre. Whether all the lymphatic glands in the body 
 
Cn. III.] OF THE BREAST. 63 
 
 participate in this function for the tissues, secreting and other, 
 with -which they are connected respective!}'', is a question that 
 lies entirely outside the present inquiry. I am bound, however, 
 to mention a particular case of the same pigmentation in lym- 
 phatic glands other than those connected with the mamma. In 
 the case of a cat that proved to have completed the involution 
 process, there were the usual evidences of pigmentation in the 
 mammae. The lymphatic glands beneath the inguinal mamma 
 were all streaked on the surface with several yellow lines, just 
 as the mesentric glands at a certain stage of digestion are streaked 
 with white lines owing to the presence of chyle in their afferent 
 vessels. The axillary glands Avere streaked in the same way. 
 The presence of pigment in both sets of lymphatic glands was 
 capable of explanation by reference to the state of the mamma. 
 But it was found, on examining the animal further, that there 
 was a chain of lumbar lymphatic glands lying along the lower 
 end of the aorta which were streaked on the surface in exactly 
 the same manner. This was at first inexplicable, and threatened 
 to upset altogether the theory of a special relation between 
 the mamma and the lymphatic glands near it. But an ex- 
 planation was forthcoming on continuing the dissection. On 
 cutting open the nterus, the mucous membrane over a small 
 area in both its horns, about half an inch from the cervix on 
 each side, was seen to have a yellowish colour. On making a 
 transverse cut at those points, a distinct deposit of a yellow 
 substance was seen in the sub-mucosa, and the microscope showed 
 this to be a collection of yellowish-brown pigment grains strewn 
 among the sub-mucous tissue, but not associated with any cell- 
 forms. Here then was an organ, known to be associated with 
 the lumbar lymphatic glands, which contained traces of pigment- 
 formation in its tissue, and which had probably been the source 
 at an earlier period of the very much larger amount of pigment 
 represented by the pigmented cells in the afferent lymphatics 
 and sinuses of the lumbar glands. The pigmentation of the 
 various sets of lymphatic glands, portal and mesentric, bronchial, 
 and others, is in all probability to be explained in a special way 
 for each of the sets. The pigmentation of the mammary lym- 
 phatic glands may be treated of by itself, all the more so as 
 the pigment is brought to them beyond doubt in the substance 
 
04 77/ A' LYMPHATIC C LANDS [Pt. T. 
 
 of cells that do Dot belong to the ordiuary circulating cells of 
 the lymph. 
 
 The lines or streaks of pigmentation on the surface of the 
 lymphatic glands in the case just referred to evidently corre- 
 sponded to the cortical lymph-sinuses, with which the in-carrying 
 lymphatic vessels are in direct connexion. In the microscopic 
 section the lymph-sinuses of the cortex are sometimes found 
 packed with the mammary pigmented cells, arranged in rows 
 and compressed into a more cubical form. From the cortical 
 sinuses they pass, like the lymph, into the centre of the gland. 
 Their progress therein, and the changes that await them, have 
 now to be traced. This part of the investigation, relating in 
 the first instance to the disposal of the cellular bye-products of 
 the mammary secretiou,'will be found to have a bearing on 
 
 The genkral Plan of Strug iure and the Function of 
 Lymphatig Glands. 
 
 The minute structure of lymphatic glands has been studied 
 by so many observers during recent years that there is little 
 room to add anything to the histological details. But there 
 need be less hesitation about offering a re- interpretation of 
 certain of those appearances ; while the whole question of the 
 function of lymphatic glands is open to elucidation even of an 
 elementary kind. As regards structure, the lymphatic glands 
 consist of three kinds of tissue, which are represented in Fig. 19. 
 The first and most abundant is the follicular, which includes 
 both the round or pear-shaped follicles proper, occupying the 
 cortex of the gland, and the cylindrical continuations of the 
 same which occupy the medullary part of the gland. The 
 follicular tissue consists of dense masses of small lymphoid 
 cells, which lie in the meshes of a fine network that can be 
 made visible by shaking the cells out of it. There is no difference 
 between the cortical follicles and the medullary cylinders as 
 regards minute structure. The second class of structures are 
 the lymph-sinuses. They accompany the follicular tissue through- 
 out the entire gland, forming somewhat narrow spaces around 
 tiie follicles at the cortex, but widening towards the medulla 
 
Cn. Ill] OF THE BREAST. 05 
 
 into spaces round the cylinders of lymphoid cells, which are often 
 as broad as the cylinders themselves. The lymph-sinuses are 
 not perfectly open and empty spaces. On the contrary, they are 
 crossed in all directions by a network of bands or fibres, and 
 they are always found more or less filled by cells. But if they 
 are compared with the follicular tissue in this respect, the 
 branching fibres within them do not form at all so close a reti- 
 culum, nor are the cells that lie among the fibres so numerous 
 or so densely packed. The lymph-sinuses, even when filled 
 with cells, have a much lighter appearance than the follicles 
 or cylinders, and they are spoken of by Recklinghausen as the 
 "lichtere Stellen^." The same writer also points out that the 
 cells in the lymph-sinuses are only temporary occupants. The 
 third kind of structure within the gland is the trabecular, of 
 which His makes a subdivision under the name of " stroma of 
 the hilus^" The ti'abecular tissue is of a densely fibrillar kind, 
 and forms the capsule of the gland at one pole, the stroma of 
 the hilus at the other, and the true trabeculge or framework of 
 the gland at all intervening points. The trabeculce are related 
 to the other structures of the gland in the manner shown in 
 Fig. 19; they divide two lymph-sinuses from each other; the 
 fibres that are stretched across the lymph-sinus are attached 
 to the follicle or cylinder on the one side of the space, and 
 to the trabecula on the other. It is now very generally admitted 
 that the trabecular tissue, including that of the capsule, contains 
 a certain quantity of involuntary muscular fibre in its substance. 
 The assertion was first made by Malpighi, on what grounds it 
 is not known^. It has been re-afiirmed of recent years by Brlicke, 
 Heyfelder, His and others on various grounds, such as the 
 isolation by means of nitric acid of highly characteristic muscular 
 fibres, with rod-shaped nuclei, and the frequent occurrence of 
 rod-shaped nuclei in situ throughout the substance of the tra- 
 beculae, Frey is of opinion that the latter cannot be distinguished 
 from young or embryonic connective-tissue cells, W. Midler, 
 after reviewinij all the circumstances, thinks the existence of 
 muscular fibres in the trabeculas to be probable, but he rests 
 
 1 Strieker's Handbueh, p. 241, 
 
 '■^ Untersuchungen ueber den Bau der Lymphdriisen, Leipzig, 18G1, 
 
 3 His, loc. cit. 
 
 c, c. 5 
 
GG THE LYMPHATIC GLANDS [Vt. I. 
 
 his opinion less on the completeness of the histological evidence 
 than on the fact made out by Brlicke and Heyfelder that the 
 l3'mphatic glands of animals can be made to contract by an 
 electrical stimulus \ 
 
 As re^rards the circulation through the gland, the afferent 
 lymphatic vessels are known to discharge themselves into the 
 lymph-sinuses at various parts of the cortex, and the outcarrying 
 lymphatic vessels to take origin from the wide lymph-sinuses in 
 the hilus. The following account of the blood-vascular distribu- 
 tion is given by His. Several arterial trunks enter at the hilus ; 
 in the stroma of the hilus takes place the coarser branching ; 
 certain of the branches pass into the trabecule ; but the greater 
 number of them pass into the meduUaiy cylinders and run in these 
 towards the periphery into the follicles. The vessel running in 
 the axis of the cylinder gives off many fine branches which open 
 into a capillary network spread out on the surface of the cylinder. 
 Out of this capillary network several small venous branches take 
 origin, which return to the hilus by a venous trunk running along 
 the axis of the cylinder and lying side by side with the artery 
 therein. As to the follicles, they receive their chief arterial supply 
 by extension of the vessels running in the cylinders, and the 
 capillary distribution and venous reflux is on precisely the same 
 system as in the cylinders. In both cases the greatest capillary 
 distribution is on the rounded surface which is directly exposed to 
 the lymph-sinus. The lymph-sinuses contain no blood-vessels. 
 
 The most recent additions to the histology of the lymphatic 
 glands concern the lymph-sinuses and their contents. The branch- 
 ing fibres that stretch across the sinuses from the follicle or 
 cylinder on one side to the capsule or trabecula on the other have 
 been usually described as branched connective-tissue cells or 
 spindle-cells containing nuclei. From another point of view all the 
 nuclear cells that are seen in the lymph-sinus have been divided 
 into two classes, those that lie free in the meshes of the reticulum 
 (the lymph-cells) and those that are the fixed cells of the reticulum. 
 Fully accepting the distinction between the two kinds of cells, 
 Bizzozero^ has endeavoured to show that the latter are not really 
 branching connective-tissue cells, but that they lie upon the fibres 
 
 1 Ilenle nnd Pfeiifer's Zeitsclirift, 1863, p. 121. 
 • Moleschott's Untersuchutigen, Vol. xi. p. 300. 
 
Ch. in.] OF THE BREAST. 67 
 
 of the reticulum in one or other of two ways: the cells of the 
 reticulum are either thrown round about the fibres, enclosing thein 
 as if in a tube of protoplasm, or they are stretched across an entire 
 mesh of the reticulum, like a window-pane in its frame. In the 
 former case the nucleus is seen to lie on the side of the fibre, in 
 the latter case it has a central position in the cell, as in a front 
 view of an endothelial cell. About the same time that Bizzozero 
 made these observations, Ranvier was also led to a somewhat 
 similar conclusion. In sections of lymphatic glands hardened in 
 picric acid solution, he was able to obtain, according to the time of 
 immersion in the solution, either the reticulum without a single 
 cellular element in or upon it, or the reticulum and its branches 
 covered by endothelial cells. He concluded that the reticulum 
 is not formed of branching cells joined together by their processes, 
 but that it is formed of fibres of the connective-tissue united in a 
 network and covered by endothelial cells \ It remains to be added 
 that W. Mliller had observed the fixed elements of the reticulum 
 to be of two kinds, mature and embryonic. The one were in the 
 form of thin connective-tissue fibres with spindle-shaped nuclei ; 
 the other consisted of a delicate finely granular ground-substance 
 (protoplasm), branching into a network, and with elliptical nuclei 
 embedded in it. Along with the elliptical nuclei there could be 
 observed other and smaller cells which represented all transition 
 stages to the mature lymph-corpuscles. Between the numerous 
 irregular openings of this network the lymph-current passed, as if, to 
 quote the apt expression of Briicke, through the pores of a sponge ^ 
 It has also been observed by Toldt that some fibres of the reticulum 
 were nucleated, while others were simply thread-like filaments ^ 
 
 Notwithstanding the numeraus histological researches on the 
 lymphatic glands, their functional import remains conjectural, and 
 in particular the elaborate filtering apparatus which is formed by the 
 branching fibres or protoplasmic bands stretched across the lymph- 
 sinuses has had no adequate use assigned to it. The observation 
 of Leydig, to the effect that "the production {Erzeugung) of 
 lymph-corpuscles is the special physiological task of lymphatic 
 
 ^ See his Traite Technique d'Histologie, p. 398. 
 
 2 W. Miiller, loc. cit. 126—128. 
 
 3 Eine Methode zur Injection der Lymphbahneu in deu Lymplidriisen. Sitzungs- 
 herichte der Wiener Akadeinie, Jan. uud Feb. 1868. 
 
68 THE LYMPHATIC GLANDS [Pt. I. 
 
 glands^" sums up pretty nearly the view that still prevails. la 
 addition to contributing to the number of lymph -corpuscles (white 
 blood-corpuscles), the lymphatic gland is supposed to purify the 
 lymph in its course through it. Thus, Virchow points out that 
 there is no free passage for the lymph through the gland ; it must 
 pass between cells more or less closely packed together. The cells 
 of the gland are like the particles of a charcoal filter; the lymph 
 percolates through them and wells out on the other side "as it 
 were filtered and purified"''," But it need not be pointed out that 
 this is only a graphic illustration, which leaves our intimate know- 
 ledge very much where it Avas before. A circumstantial account of 
 the physiological mechanism of the gland is given by His. The 
 lymph entering the gland by the afferent lymphatics circulates 
 under low pressure in the lymph-sinuses and moves slowly therein 
 towards the hilus. The pressure is insufficient to drive the lymph 
 from the lymph-sinus into the gland-substance. On the other 
 hand, the blood-pressure will cause a stream to flow from the 
 highly vascular gland- substance outwards to the sinus, and this 
 stream will carry with it, along with the fluid transuded from the 
 capillaries, a number of the lymphoid cells that are being constantly 
 reproduced in the gland-substance, and so convey them to the 
 lymph-stream. With the contraction of the capsule and trabeculas, 
 (a rythmical contraction depending on the presence of involuntary 
 muscular fibres in them,) the whole gland seeks to diminish in 
 size, and as the afferent lymphatic vessels possess valves to check a 
 reflux, the lymph-sinuses will empty themselves into the outcarry- 
 ing lymphatic vessels. At the same time, in consequence of this 
 contractile pressure, the gland-substance will for the moment 
 contain less blood. As soon, however, as the contraction ceases, it 
 must happen, in consequence of the sudden relaxation of tension 
 in the interior of the gland, not only that the gland-substance will 
 become again richer in blood, but also that the emptied lymph- 
 sinuses will again become full. The re-filling of the lymph-sinuses 
 may take place both from the afferent lymphatics and from the 
 gland-substance ; if the afferent lymphatics contain no lymph, then 
 the lymph-sinuses are filled entirely by the fluid that has transuded 
 from the gland-substance'. One function of the lymphatic glands 
 
 1 Lehrbuch der Histuloyle,lS57. 2 CeUular-rathohigie, Tp. 208. 
 
 3 Loc. cit. p. 21. 
 
Ch. Ill] OF THE BREAST. 69 
 
 appears then to be tliat of adding to the lymph-corpuscles of 
 the circulating fluid a certain number more of the same kind 
 derived from the follicular tissue of the gland. Recklinsfhausen 
 observes that the multiplication of lymphoid cells in the follicles 
 has in fact never been seen, but that it is indirectly probable 
 for several reasons, the chief of these being that the lymph 
 leaving the gland contains more corpuscles than the lymph enter- 
 ing it. 
 
 Returning to the material of this research, there are introduced 
 factors of a new order altogether, which may or may not be special 
 to the lymphatic glands of the mamma. We have to deal with 
 the conveyance to the lymphatic gland of a peculiar class of cells, 
 which are waste products of the secreting gland ; and we are 
 concerned not so much with the production of actual new cells 
 within the lymphatic gland, as with the changes that it effects 
 upon a certain quantity of crude cellular material supplied to it 
 from without. Now, it is in this altered stating of the problem 
 that we may find some light thrown not only on the function of 
 lymphatic glands but also on their structure. It has been already 
 mentioned that the large yellow granular cells which are produced 
 within the mammary acini in certain states of the function, and 
 are often to be seen in the tissues outside, are also found in the 
 lymph-sinuses of the lymphatic gland. It is not impossible that 
 an objection may here be taken as to the actual identit}'' of the 
 cells found in the two situations. It may be said, for instance, 
 that the large yellow cells lose their pigment within the gland, as 
 no doubt they sometimes do, and that the pigment-granules are 
 taken up by the ordinary lymph-corpuscles that come in contact 
 with them, and so get conveyed to the lymphatic gland. It is not 
 unreasonable to suppose that some such explanation may apply to 
 the pigmentation of lymphatic glands in other situations, especially 
 where the pigment seems to come from red blood-corpuscles. 
 But in the case of the mammary lymphatic glands, there can be no 
 doubt that the large yellow cells that originate in the secreting 
 structure pass bodily into the lymphatic gland. Entire cells, more 
 than twice the size of lymph-cells, are found in the lymph-sinuses, 
 and the collections of them in the lymph-sinuses have often a 
 close resemblance to the collections of the same cells in the inter- 
 lobular tissue. But in the lymph-sinuses the pigmented cells of 
 
70 THE LYMPUATIC GLAXDS [Pt. I. 
 
 the mamma begin to undergo changes of form and reductions in 
 size, whereby their identity becomes lost. 
 
 When a somewhat loosely packed lymph-sinus is examined in 
 one of the pigmented mammary lymph-glands, it becomes at once 
 apparent that the pigmented waste-cells occupy the place of the 
 branched cells of the reticulum; their nuclei are the nuclei of the 
 reticulum, and their pigmented bodies and processes lie upon the 
 expansions or branchings of the same. Fig. 18 on a later page is 
 a representation of this appearance. There are to be seen in the 
 reticulum no other cells or nuclei but the pigmented cells and 
 their nuclei ; the only other elements of the reticulum are the 
 fibres. These are fixed structures in the lymph-sinuses, and in 
 the preparations that are now being described they appear to be the 
 only fixed structures. I am, however, not in a position to deny 
 the existence of the endothelial cells described by Ranvier, which 
 require special methods of preparation to make them visible. At 
 the same time, if all the cells of the reticulum can be equally 
 dislodged, it seems to me somewhat arbitrary, in the absence of 
 very specific differences, to describe some of them as endothelial 
 cells and others as cells belonging to the circulating lymph ; and 
 as regards the cells described by Bizzozero, cells that were wrapped 
 round the fibres like a sheath of protoplasm, or that stretched 
 across the meshes like the panes of glass in a window, I have no 
 doubt that they correspond to the pigmented cells that I now 
 describe, and that they belong to the circulating cells. The real 
 distinction among the cells in a lymph-sinus is that some of them 
 adhere more closely than others to the fibres of the reticulum ; 
 and the significance of those differences among the pigmented 
 waste-cells has now to be considered. 
 
 Many of the large yellow granular cells in the lymph-sinus are 
 of their usual round form ; but many of them — and in the emptier 
 lymph-sinuses the whole of them — are either drawn out into 
 processes which follow the direction of the fibres, or they are other- 
 wise stripped of their protoplasmic cell-substance and their pig- 
 ment, in all degrees down to the condition of naked nuclei. The 
 various steps of the process by which the cells lose their protoplasm 
 and their pigment in the meshes of the reticulum are so clear that 
 the fact in its entirety becomes well established. For example, a 
 large yellow granular cell with visible nucleus has its substance 
 
Ch. III.] OF THE BREAST. 71 
 
 drawn out into two or three or four processes which correspond to 
 the direction of fibres passing across the sin\is. Again, a row of 
 free pigment-granules is seen to be the continuation of a somewhat 
 truncated cell-process, the granules lying in or upon the fibre. 
 Again, the pigmented cell is found stripped of all its cell-substance 
 except a mere yellow fringe round the nucleus. Lastly, from 
 another point of view, nuclei are constantly met with which have 
 a few granules of yellow pigment clinging to them. One might 
 multiply indefinitely such examples of transition appearances in 
 the large pigmented cells while they are in the course of losing 
 their pigmented cell-body. 
 
 The next question that arises is the manner in which the denu- 
 dation takes place, or the part that the so-called fibres of the 
 reticulum play in the same. Reference has been made to the 
 distinction by W. Mliller of two kinds of substance occurring as 
 the reticular meshwork. One kind he describes as the ordinary 
 connective-tissue bundles containing the ordinary spindle-cells ; 
 the other kind was a finely granular protoplasmic substance which 
 contained large elliptical nuclei, and broke up into branches. 
 Along with the elliptical nuclei, he found in the protoplasmic sub- 
 stance a number of smaller cells showing all gradations from the 
 large elliptical cells down to the ordinary lymphoid cells of the 
 follicles. The small cells he attributed to a proliferation of the 
 nuclei within the protoplasmic substance, and he figures certain 
 appearances of clusters of nuclei which seemed to point to a 
 process of cell-division. The same protoplasmic condition of the 
 reticulum and the same varieties of cells in its substance are also 
 recorded by Frey in his text-book, and the figure given by the 
 latter (Fig. 389) is entirely confirmatory of the view that is now 
 to be stated of these appearances. Frey mentions incidentally 
 that he had seen the same protoplasmic meshwork in the lymph- 
 sinuses of the mesenteric glands, and that he had seen the fat- 
 molecules of the chyle within the bands of protoplasm; and he 
 adds that he then made the mistake of supposing the reticulum 
 to be " a system of cellular tubes." The true explanation seems 
 to be not that the reticular substance is a fixed system of tubes, 
 but that it can be made tubular by cells passing into it. There 
 is often to be seen, in the course of a fibre, the appearance of a 
 sudden hollow expansion within which lies a nucleus. The out- 
 
72 THE LYMPUATIC GLANDS [Ft. I. 
 
 lines of the expansion usually taper abruptly, and become con- 
 tinuous again with the narrow fibre beyond ; but sometimes the 
 breadth of the fibre beyond the nucleus is hardly less than at the 
 point of expansion. In one of the preparations of the series 
 there was seen a long " fibre " of the lymph-sinus which had no 
 fewer than three separate dilatations on it, each containing a 
 round nucleus, while there seemed to be a free passage in its sub- 
 stance between them. The appearance was not unlike the beaded 
 condition of a double contoured nerve-fibre. Whatever may be 
 the exact process, there is no doubt that the large yellow granular 
 cells become entangled in the meshwork of the lymph-sinus, and 
 thereby lose their pigmented protoplasm. In all probability they 
 travel through a considerable length of sinus, and come in contact 
 with the reticulum many times before they are reduced to their 
 mi clear condition. The disposal of them after this point falls to 
 be considered separately. The use of the lymphatic gland may 
 be compared not so much to that of a charcoal filter, as to that 
 of a winnowing apparatus. In the first step of the process the 
 husk is loosened and detached from the grain or kernel, and in 
 the second step of the process the useful elements are separated 
 and collected together. In addition to those changes, which are 
 effected within the lymph-sinuses, it seems probable that the cells 
 passing through the lymphatic gland sufier further changes in 
 entering the follicular tissue. 
 
 In the accounts of the involution and evolution process it was 
 stated that the large yellow granular cells were sometimes reduced 
 to the nuclear condition, with loss of their pigment, while they 
 were still in the mamma. Collections of them can be seen in the 
 fibrillar tissue which show very much the same gradations from the 
 full-sized nucleated granular cell to the naked nucleus, as has 
 been described for those in the lymph-sinus of the lymphatic 
 gland. Again, in the final state of involution, or the resting state, 
 the contracted acini are found to have a few grains of pigment 
 lying free among the resting cells, and in some cases where the 
 free pigment is not seen in the acini it may be seen strewn over the 
 surface of the lobule generally. Now, within the lymphatic gland, 
 the protoplasmic reticulum of the lymph-sinus is not the only 
 agency for reducing the granular cells to their nuclear condition. 
 Fig. 18 is a drawing from the case in which the lymphatic glands 
 
Ch. Ill] OF THE BREAST. 73 
 
 beneath the mamma were in a rudimentary condition; that is, the 
 follicular tissue was at its lowest point, while the lymph-sinuses 
 
 Fio IS. 
 
 3." fi 
 
 From a lymphatic gland beneath the mamma. The follicular tissue 
 rei)reseiited by two or three rows of lymphoid cells in the adventitia 
 of a lar^e vein; the lymph-sinus occupied by large' yclhno cells with 
 their substance often drawn out into processes. Magnified 300 diameters. 
 
 and the stroma of the hilus were of great extent. The figure 
 shows the section of a large vein, and the follicular tissue is 
 represented by two or three rows of lymphoid cells in its adven- 
 titia. Surrounding the vein with lymphoid cells in its adventitia 
 is the large lymph-sinus filled with pigmented cells, many of 
 which are drawn out into processes, as already described, while 
 some of them have their cell-substance reduced in amount. It 
 will be observed that the lymph-sinus is separated from the rows 
 of lymphoid cells by a membrane, which is, as it were, the outer- 
 most layer of the loose adventitia of the vein; it appears in the 
 figure as a dark line. This limiting membrane showed in the 
 coloured preparation an interesting condition of things which 
 could not very well be transferred to the figure. It showed at 
 one or two points a pigmented cell, as if adhering closely to it, 
 drawn out to a long narrow form, and with its pigmented granules 
 scattered along the membrane for a considerable distance on each 
 side of the cell-mass. The cell was no doubt passing from the 
 lymph-sinus into the follicular area, and losing its pigmented 
 substance in the process. In the preparations of the same case 
 another very definite appearance is often to be seen. In parts 
 
74 THE LYMPHATIC GLANDS [Pt. I. 
 
 of the gland where the follicles are of considerable size, the 
 surface of the follicle is strewn with numerous grains of pig- 
 ment approaching to red in colour. These grains are sometimes 
 of considerable size, circular, and with the characteristic dark 
 border. The lymph-sinus surrounding the follicle is full of the 
 ordinary yellow nucleated cells ; the pigment of the latter is much 
 paler than that strewn over the follicle. The collections of the 
 darker coloured free pigment are limited exactly to the follicular 
 tissue. The inference seems to be that some of the cells enter 
 the follicles with their pigmented substance, or at least a portion 
 
 Fig. 19. 
 
 / 
 
 fj 
 
 
 
 
 ^. /^^^^^ ^^ 
 
 ^3?^o I - ^l^^ >^gSs 
 
 
 
 
 C¥,^ )oc 
 
 ^^ 
 
 Cr 
 
 
 From tho medullary part of a lymphatic pland beneath the mamma of a cat: a, trabrcula 
 crowded with spindle-cells, some of which have the same substance as the larf^e dark cells in 
 the lymph-suiuses {0); c, portion of a lymphatic cylinder. Magnitied 3"0 diameters. 
 
 of it, still adhering, and that the pigment which they lose there is 
 not at once carried away, as it would have been if the denudation 
 had taken place in the sinus through which the lymph-current is 
 passing. There is one other mode of reducing the granular cells 
 
Cir. III.] OF THE BREAST. 75 
 
 to a nuclear state, which altliough it may be exceptional, can 
 hardly be passed by without mention. It is represented in Fig. 19, 
 which has been drawn as accurately as possible. The lymphatic 
 glands are those from beneath the mamma in the same case from 
 which Fig. 14 is taken. The latter shows abundant production of 
 the yellow granular cells in the mamma. Now the lymphatic glands 
 did contain traces of yellow pigment, but the large nucleated 
 granular cells that were found in the lymph-sinuses were not often 
 of a yellow colour, but they usually stained very dark by the 
 colouriug reagent (logwood). Their granular character is more 
 distinct than in the case of the yellow cells, and they have often a 
 mulberry-like surface. In other respects they are exactly the same 
 as the large yellow cells. They have been found in the mammary 
 lymiDhatic glands of two other cases somewhat more advanced 
 in evolution, and in the particular case described already, of a 
 lymphatic gland made up almost entirely of lymph-sinuses and 
 pigmented cells in them, they were found here and there among the 
 ordinary yellow cells. There is no doubt that they are only a modi- 
 fication of the large yellow cell as regards its pigment, or in the 
 property of taking on an artificial colour. The portion of gland 
 drawn in the figure is taken from medulla not very far from the 
 hilus. Portions of three cylinders of the medulla occupy the outer 
 parts ; next to them come the corresponding lymph-sinuses, and 
 in the centre running horizontally, is a broad trabecula with an 
 extension upwards. The lymph-sinuses show indistinctly the fibres 
 of the reticulum, and they contain, besides a number of the large 
 granular cells, many nuclear cells of various sizes. If the trabe- 
 culse are examined, they are found to present a very singular 
 appearance. They are crowded with staff-shaped or spindle-shaped 
 cells (or nuclei), which follow the direction of the trabecular 
 fibrils with their long axis. About one-fourth of the elongated 
 bodies have the intensely dark granular appearance of the large 
 round cells in the sinus, and these are also the largest and broad- 
 est ; the majority of the elongated elements are narrower, pale 
 and granular, and there are transition stages as regards depth of 
 staining and shape between those two kinds. The contention is 
 fhat the elongated cells in the trabecula have penetrated from the 
 lymph-sinus, and have acquired their shape in passing through the 
 narrow interfascicular spaces. No other explanation of this picture. 
 
76 THE LYMPHATIC GLANDS [Pt. I. 
 
 as a whole, suggests itself. The staff-shaped nuclei, if not the 
 spindle-shaped, have often been seen in the trabeculae of lymphatic 
 glands, tltough they do not appear to have been figured before in 
 such numbers or in such variety. The usual account of them is 
 that they are the nuclei of involuntary muscular fibres ; but it has 
 been pointed out already that some authorities are not satisfied 
 with the histological evidence for this allegation, and that W. 
 Miiller, in weighing the evidence, is led to accept the theory chiefly 
 because the lymphatic glands have been shown to contract under 
 an electrical stimulus. 
 
 The presence of plain muscular fibres in the trabecular and 
 capsular tissue of lymphatic glands is a matter that does not in 
 any way concern the general purpose of this investigation ; and if 
 those fibres have actually been isolated by means of dilute nitric 
 acid in " highly characteristic " forms, it is futile to offer any other 
 exclusive or dogmatic explanation of the appearances in the trabe- 
 culae as they are seen in sections. At the same time I do not 
 doubt that Fig. 19 represents a crowd of circulating cells passing 
 through the substance of the trabeculas, instead of or in addition 
 to passing through the meshwork of the sinus. The difference 
 between the trabecula and the meshwork of the sinus is not so 
 very great as it appears at first sight. The "fibres" or protoplasmic 
 bands crossing the sinus are simply bundles detached from the 
 side of the trabecula ; the reticulum of the lymph-sinus is a 
 frayed-out condition of the trabecular tissue. By following the 
 edge of a trabecula it is easy to see " fibres " detaching them- 
 selves from it, or to trace the continuity between fibres of the 
 sinus and the outermost bundles of the trabecula. The fibre 
 usually crosses the sinus from the trabecula in such a way as to 
 make a small angle with the latter, and a circulating cell of the 
 lymph that had been driven into that angle would, in its further 
 progress, necessarily enter the substance of the trabecula. There 
 are appearances in the preparation from which Fig. 19 is taken 
 that seem to bear out that notion. Again it is not to be forgotten 
 that the cells or nuclei in the trabecuki3 vary very much in 
 different cases, not only in number but also in shape — many of 
 them being round — and that circumstance is hardly consistent with 
 the theory that they are all the nuclei of plain muscular fibres. 
 
 The correctness or incorrectness of the explanation given of 
 
Ch. III.] OF THE BREAST. 77 
 
 the staff-shaped and spindle-shaped cells in the trabeculse in 
 Fig. 19 has a bearing on the question already raised with reference 
 to Fig. IG. The latter is held to show a particular form of the 
 Avaste products of evolution on their way through the interfascicular 
 spaces of the mammary interlobular tissue, and the exactly analo- 
 gous occurrence in the trabecular tissue of the lymphatic gland 
 may be lield to give an independent support to that view. 
 
 Up to this point we have found one function at least of the 
 mammary lymphatic glands to be the reduction of the granular 
 waste-cells of the secretion to the status of nuclei. We have 
 seen that such denudation takes place chiefly among the proto- 
 plasmic fibres and bands stretched across the lymph-sinuses, 
 but that a similar change is effected on the large waste-cells 
 in their passage through the limiting membrane of the follicles 
 or cylinders, or in their passage through the interfascicular spaces 
 of the trabeculse. The special use or significance of the follicular 
 tissue of the gland has not yet been touched upon, and although 
 that is a more difficult point than the preceding, some of the 
 preparations of this series are not without instruction as to it. 
 The Fig. 20 is taken from a case already referred to. The lym- 
 phatic glands were quite small bodies in the midst of the cushion 
 of fat beneath the mamma in the inguinal region. They seemed 
 to the naked eye to be composed almost entirely of yellowish 
 pigmented material, which was found to be the waste material 
 of the secretion ; and it appeared as if the conveyance of so 
 large a quantity of waste-cells to a particular point in the fat- 
 tissue beneath the mamma had given occasion for a lymphatic 
 gland to develope itself, its framework being derived from certain 
 fixed elements of the part — blood-vessels and stroma — and its 
 parenchjmaa being formed by the waste material in its crude 
 or in its subsequent transformed state. Without raising the 
 question, whether a set of lymphatic glands may not have at 
 one time existed at that part of the fat-tissue only as an empty 
 framework, it does appear from the examination of the glands 
 as they were found, that their chief and earliest constituents 
 were the dense and abundant stroma of the hilus, the larefe 
 lymph-spaces, and the blood-vessels. The glands were in fact 
 rudimentary glands, and that not in the sense of embryonic 
 development, but in the sense of a disused function being suddenly 
 
78 
 
 THE LYMPHATIC GLANDS 
 
 [Pt. I. 
 
 revived. The animal Avas aged ; and tlie great production of 
 waste-cells, though it followed in the track of the evolution of 
 
 Fio. 20. 
 
 
 ^.m 
 
 ■"^^K 
 
 
 
 
 
 f 
 
 From a l.ymphatic plaiid (same oasp as Fip. IS). In the centre of the figure is the 
 loiigitvidinal section of h vein wliicli discharKes below into a larger vein; Ilie 
 follicular tissue is represented liy rows of lymphoid cells in the adveiititia of the 
 vein, and especially of its branches; the l^uiph-sinuses are tilled by lar//e yellow 
 granular cells. Magnified 150 diameters. 
 
 the breast, was not due to the ordinary cause of evolution, but 
 to some unknown pathological excitation. If the drawing be 
 
Cir. III.] OF THE BREAST. 79 
 
 referred to, it will be seen that the dense fibrillar stroma of tlie 
 liilus is the most abundant tissue ; that it is penetrated in various 
 directions by wide lymphatic spaces, the lymph-sinuses; and 
 that the follicular tissue is represented only by a few rows of 
 lymphoid cells in the walls of the veins. One of the latter is 
 shown in longitudinal section, with several branches. Beginning 
 at the top of the figure, there is seen at the left-hand corner a 
 considerable follicle or cylinder, in the centre of which is a vein. 
 On the other side at a corresponding point, there is another cross 
 section of a vein with a smaller number of lymphoid cells round 
 it, and if the longitudinal vein be traced downwards, it will be 
 seen that the lymphoid cells which form two or three rows at 
 its upper part and round its branches, diminish in number at 
 its lower part, and, where the vein opens into the large venous 
 trunk at the bottom of the figure, that they are altogether 
 wanting. The figure is nothing less than a diagram of the plan 
 of a lymphatic gland on a small scale. The blood-vessels have 
 long been known to run in the axis of the lymphatic cylinders, 
 and several writers^ have suggested what is undoubtedly the 
 true account of the relation of the lymphoid corpuscles to them. 
 They have described the cylinders of the gland to be simply the 
 loose adventitia of the blood-vessels containing lymphoid cells. 
 This view of the matter at once places the follicular tissue of 
 the lymphatic glands in the same category as the follicular tissue 
 of the spleen. The Malpighian bodies seated on the fine arterial 
 twigs are to be considered as true lymph-follicles ; they form 
 circumscribed thickenings on the walls of the arteries, and are 
 to be viewed as simple deposits of colourless cells (lymph-cells) 
 in the interspaces of the adventitia. In many animals the 
 alveolar thickenings are not circumscribed, but are more uni- 
 formly distributed over the arterial wall. It is to be noted 
 that in the spleen the collection of lymphoid cells takes place 
 in the walls of the arteries, and that in the accounts of lymphatic 
 glands, the arteries and veins are not spoken of separately, but 
 the blood-vessels in general. In all the preparations of the 
 above case it appeared to be the adventitia of the veins in which 
 the lymphoid cells were collected. Such, then, appears to be 
 
 1 Leydig, loc. cit. ; Billroth, Beitriiije zur patliol. Histologic, Berlin, 1858. 
 
80 THE LYMPHATIC GLANDS [Pt. I. 
 
 the relation of the trunks of the vessels to the lymphoid cells in 
 the centre or hilus of the gland ; the cylinders of the medulla are 
 essentially their loose adventitia filled with lymphoid cells. When 
 we come to consider the follicles of the cortex, a corresponding 
 explanation of them may be given. The medullary cylinders 
 contain all the larger blood-vessels, which are provided with 
 stout adventitia ; some of their branches break up into capil- 
 laries on the cylinders, and the rest proceed to the cortex of 
 the gland. As the vessels divide into smaller branches, the 
 looser does the connective tissue of the adventitia become, and 
 at length there appear, in the place of connective-tissue fibres, 
 '■ cytoblasts " whose cell-membranes enclose narrow spaces, and 
 which are provided with two or three thread-like processes \ 
 It is in that reticulum that the lymphoid cells lie, generally 
 two or three of them in one mesh. The small-meshed reti- 
 culum of the follicles is simply a modification of the adventitia 
 of the terminal vessels. At the same time we have seen in 
 these preparations (Fig. 20) that the smaller the vein is, the 
 larger is the collection of lymphoid cells in its wall. As the 
 cortex or periphery of the gland is approached, one finds that 
 the adventitia of the vessels is frayed out into a spacious mesh- 
 work, and that the lymphoid cells have proportionately increased 
 in number. 
 
 Some facts have been mentioned already to show that the 
 waste-cells are sometimes denuded of their substance and pig- 
 ment in passing from the lymph-sinus into the wall of the vein, 
 i.e. into the follicular tissue ; and it is probable that the cells that 
 have been reduced to the nuclear condition in the lymph-sinus 
 itself, or in the substance of the trabeculte, enter the cylinders 
 and follicles in the same way. This brings us to consider what 
 is the further effect of the lymphatic gland on the waste-cells 
 brought to it, besides the two uses of it which have been com- 
 pared to the uses of a winnowing apparatus. The lymphatic 
 follicles no doubt represent the heaps of cells that have been 
 cleared of their pigmented and granular cell-substance ; the 
 follicles are a sort of reservoir in which they collect, and from 
 which they may be again swept out into the lymph-current, in 
 the manner described by His in a passage already quoted. But 
 
 1 Briicke, Vchcr die Chylusgefdsse und die Resorption des Chylus, p. 34, 35. 
 
Ch. III.] OF THE BREAST. 81 
 
 they have evidently another purpose. The meshwork of the 
 follicle differs considerably from that of the lymph-sinus ; the 
 meshes are very much finer, and they are of a uniform size. 
 If the lymph-sinus in Fig. 19 be examined closely, it will be 
 seen to contain some very large nuclei, and some of the ordinary 
 lymphoid size; the large nuclei have a distinct linear contour, 
 a pale homogeneous substance, and a comparatively large nu- 
 cleolus. Their origin in this particular case belongs to a former 
 part of the inquiry ; they are probably the characteristic waste 
 products of the middle period of evolution. It may be said of 
 lymphatic glands in general that the cells in the lymph-sinus 
 are not only fewer than those in the follicles and cylinders, but 
 tliat they are also a size larger and that their substance is paler- 
 There appears to be sufficient evidence, more especially in the 
 remarkable case from which Fig. 18 and Fig. 20 have been 
 taken, that cells do leave the lymph-sinuses and enter the folli- 
 cular tissue. In so doing, they would pass from a network of 
 large and irregular meshes to a network of smaller meshes of a 
 uniform size. Those differences in the reticular tissue of the 
 two situations correspond to differences in the cells that re- 
 spectively occupy them ; the cells of the lymph-sinus are of 
 various sizes and on the whole large, while those of the follicular 
 tissue are of a uniform smaller size. The conclusion seems 
 justified in the case of the cells brought to the lymphatic glands 
 from the breast, that as the network of the lymph-sinuses strips 
 some of them of their pigmented cell-substance, so the follicular 
 meshwork may reduce others of them to the uniform size of 
 lymphoid cells. 
 
 We have thus followed, step by step, the evidence to show 
 that the waste-cells of the mamma enter the special lymphatic 
 glands of the organ, that they undergo various degrees of change 
 according to their requirements, and that they do not leave 
 the gland till they have acquired a uniform size. The finishing 
 touch of their reduction takes place in the follicles, and from 
 these they issue on a new career. All that has been asserted 
 of the follicles as matricular organs for the formation of lymphoid 
 cells may be asserted of them, mutatis mutandis^ as organs for 
 the final moulding of the same. 
 
 The utilisation of the waste products of the mammary secre- 
 c. c. 6 
 
82 THE LYMPHATIC GLANDS OF THE BREAST. [Pt. I. 
 
 tion naturally implies a large addition, present or future, to the 
 number of white corpuscles in the blood ; and such an increase 
 is found to take place during pregnancy. This subject attracted 
 the attention of Virchow at an early period and is discussed in 
 his Gesammelte Ahhandlungen. In the Cellular-Pathologie he 
 describes it as a case of physiological leucocytosis. In his earliest 
 notice lie pointed out that those periodical changes of the blood 
 did not take place within the blood itself; the cause of them 
 lay without, and he specially mentions " organs of secretion and 
 reception, but especially the proper matricular organs of the 
 blood, among which the lymphatic glands must be placed first\" 
 In the Cellalar-Patliologie the subject is noticed as follows : 
 " In proportion as a pregnancy advances, as the lymphatic vessels 
 of the iiterus enlarge, as the Stuffwechsel in the womb increases 
 Avith the development of the foetus, the lymphatic glands of 
 the inguinal and lumbar regions become considerably enlarged, 
 and sometimes to such an extent that if we found them at 
 any other time Ave should say they Avere inflamed. This en- 
 largement brings to the blood more and more new particles of 
 a cellular nature, and so the number of colourless corpuscles 
 rises from month to month"." The pelvic region must no doubt 
 be held to be the chief source of the physiological leucocytosis 
 of pregnancy, but there is every reason to believe that the 
 abundant cellular Avaste products of the breast contribute also 
 to that condition. As to the fact of a leucocytosis in the animals 
 used in the study of the evolution of the mamma, the observations 
 Avere not conclusive. The point Avas neglected in the earlier 
 cases, and in only tAvo of the later Avas the increase of Avhite 
 corpuscles very marked. In both of these there Avere from 30 
 to 50 in the field of the microscope ^ 
 
 1 Loc. cit. p. 760. 2 Cdlular-Tathologie, p. 229. 
 
 3 There is another class of appearances in pregnancy, besides the lencocvtosis, 
 on which the study of the evolution of the mamma throws some light. These are 
 the peculiar deposits of pigment. Those of them that occur in women along the 
 lower part of the hnea alba, are perhaps to be referred to a corresponding local 
 source in the pelvis. The pigmentation of the areola may be referred directly to 
 the pigment production within the mamma during evolution. It is obvious that if 
 the pigmented cells in Fig. 13 travelled along the submucosa of the ducts all the 
 way to their orifice, instead of breaking into the lumen, they would form a deposit 
 of pigment in the areola; and that may reasonably be supposed to be the track 
 along <v\hich the pigment is carried, even if there is no actual transport of the 
 original pigmented cells. 
 
CHAPTER IV. 
 
 THE DEVELOPMENT OF THE BREAST. 
 
 The opinion that is at present generally accepted on the develop- 
 ment of the breast is that it is a complex extension downwards of 
 the ectoderm. From observations made on the human foetus, it is 
 said to begin, in the fifth month of foetal life, as a solid rounded 
 process of the rete mucosum ; before the seventh month, this wart- 
 like body has thrown out a number of offshoots from various points 
 of its circumference. From that rudiment the whole complex struc- 
 ture of the gland is said to grow out by a process of continuous 
 extension. The rudiment of the gland is at first solid, and the 
 excavation of cavities throughout the various extensions of the 
 rudiment is associated with the formation of fat-containing cells in 
 their interior. These fat-containing cells, by means of Avhich the 
 channels of the gland are formed, constitute the milk of the new- 
 born, which is therefore a developmental product^ 
 
 The development of the breast is thus held to be a process of 
 budding from the mucous layer of the skin, and the homology of 
 the organ is always sought for in other accessory cutaneous glands, 
 and chiefly in the sebaceous glands. The acquisition of the 
 mamma as a mammalian characteristic has also been discussed by 
 distinguished writers in a speculative way from the same point of 
 view; it has been supposed that external agencies acting upon the 
 skin-glands of a particular region have modified them in such a 
 way as to form the milk-gland. 
 
 1 Kolliker, Entwichlunjsgeschiclite, 1st Edition, p. 3-i5. 
 
 G— 2 
 
84 THE DEVELOPMENT [Vi. I. 
 
 It will appear, however, on inquiring into its source, that the 
 notion of the mamma being a continuous extension downwards of 
 the skin, and the important doctrine of homology that is based 
 thereon, are supported by evidence of a fragmentary and more or 
 less casual kind, which was probably never intended to be accepted 
 as final. The first statements on the development of the breast 
 appear to have been made by Kolliker in an early edition of his 
 Geivehelehre, and are substantially the same as those already 
 quoted from his Entiuicldungsgeschiclite. Every one who is familiar 
 Avith the text-books of that author will recognise the great services 
 that he has rendered to histology and development V)y filling up 
 from his own observations many of the blanks that a systematic 
 writer on those subjects is apt to encounter; and his account of 
 the development of the breast must be judged as one of those 
 occasional and sometimes incomplete descriptions. It has, how- 
 ever, had the fate to determine the character of all subsequent 
 opinion on that subject down to the present time. Within a short 
 time of Kolliker's first reference to the subject, a memoir was pub- 
 lished by Langer on the human breast, which included an account 
 of the development. The conclusion of Langer is, that " the first 
 development of the mammary gland is bound up with the exist- 
 ence of a peculiar and independent body {an die Existenz eines 
 eigenthumlichen, selhstdndigen Koi^pers gehunden), in which the 
 ducts develope without connexion with the groove in the skin ; 
 the ducts are therefore not invaginations or processes of the outer 
 skin (somit keine Ausstvdpungen oder Fortsetzungen der dussern 
 Haut sind^.") It appears to me that that conclusion differs from 
 the view of Kolliker in essential points; but in the work of Remak, 
 which holds a position of authority in the literature of develop- 
 ment, the two authors, Kolliker and Langer, are curiously enough 
 quoted together and without distinction as having shown that 
 "the mammary glands, like the sweat glands, are formed from the 
 Malpighian layer of the upper skinl" There has thus arisen an 
 apparent unanimity as to the development of the mamma by in- 
 growths from the ectoderm, and that view still passes unquestioned. 
 
 The plan according to which most of the secreting glands of 
 
 ^ Denkschriften der Wiener Akademie, 1852. 
 
 " Untersuch. iiher die Entwickl. der Wirhdthlcrc, p. 100. 
 
Cii. lY.] OF THE BREAST. 85 
 
 the body, including tlie breast, are supposed to develope, is that 
 the rudiment coniniences as a downward process of the epithelial 
 surface on which the gland, in its mature state, will pour its secre- 
 tion. The gland is essentially a very complex reduplication of the 
 epithelial surface S The development of the gland, according to 
 this view, may also be said to be centrifugal ; the growth takes 
 place by the throwing out of processes from a central point of the 
 surface towards a wide periphery at the depth. The botanical 
 terms of "bud," "offshoot," &c., are constantly used to describe the 
 process of formation of a racemose gland. The structural growth, 
 however complex, is held to be one of continuous extension down- 
 wards, just as the growth of a plant is an extension upwards. The 
 ducts of a gland grow like the branches, and the acini unfold like 
 the leaves. But it must be held at the outset to be an abuse of 
 terms to transfer, in so facile a manner, the language of plant- 
 growth to the formation of a racemose organ within the body. It 
 is only under exceptional circumstances, as for instance in watching 
 the formation of blood-vessels in the tadpole's tail, that one can 
 observe the actual process of extension; and there is reason to 
 think that writers on development have not always respected the 
 difference between the facts derived from immediate observation 
 and the facts derived from mediate evidence. In the microscopic 
 examination of dead tissues one never observes anything but co- 
 existences, but in descrif>tions we use the language of succession ; 
 and, even if no fallacies have crept in through such laxity of ex- 
 pression, the evidence that is needed to make the mediate truth of 
 the same value as the truth of direct observation is apt to be 
 deficient. The conclusions that I shall endeavour to make good 
 in this chapter are radically opposed to the curreQt view of gland- 
 development. If the development of glands, as generally g,ccepted, 
 be called a centrifugal process, the process that I shall describe 
 for the breast may be called centripetal ; the secreting structure 
 is formed in a layer (of mesoblast) beneath the outer skin, and the 
 ducts or communications with the skin are essentially secondary 
 formations owing their existence to a force from within. The same 
 
 ^ "The plan of augmenting the secreting surface by recession or inversion of 
 the membraue, iu form of a cavity, is, with few exceptions, that generally adopted 
 in the construction of secreting glands." Quain's Anatomy, 8th Ed. Vol. ii., 
 p. 231 
 
86 THE DEVELOrMEN'T [Pt. I. 
 
 principle may or may not extend to other secreting glands ^ ; the 
 breast is at least one of the most favourable cases in which to 
 exemplify it, from the circumstance that it developes late in foetal 
 life and in a region where the surface-cells are at that period 
 easily distinguished from the deeper tissues. 
 
 Although the doctrine of continuous extension from a central 
 point on the surface has been practically the only one hitherto 
 admitted, an entirely different account of gland-development had 
 been put forward in 1842 by Good sir. Those remarkable and far- 
 reaching views of Goodsir can hardly be said to have gained cur- 
 rency, perhaps owing to the aphoristic manner in which they were 
 expressed. They are for the most part stated in naked propositions, 
 without the support of the laborious details on which they had no 
 doubt been based. Goodsir's statements are as follows ^ : 
 
 1 A somewhat similar view, with an important reservation as to the cliacts, has 
 Jately been put forward by Schenk for certain other glands. In an investigation on 
 the development of the pancreas, he draws a distinction between the source of 
 the secreting cells of the organ, and the epithehal lining of the ducts. He derives 
 the former from the mesoblast, and the latter from the hypoblast (Die Bauch- 
 speicheldriise des Embryos, Anat. u. Flnjsiol. Untersuch., quoted in Foster and 
 Balfour's Emhryolofjy , p. 133). In his Lchrhuch der Verglcichenden Embnjologie 
 der Wirhelthiere, the same author states this opinion more generally. Speaking 
 of the development of glands connected with the epithelial-glandular layer he 
 says: "Wir miissen hier gleich anfangs darauf aufmerksam machen, dass man 
 die Anlage der sogcnannten Darmdriisen, deren Entwickluug nach Eemak im 
 inneru Keimblatte stattfindet, nicht im Sinne der Jilteren Autoren auffassen kann, 
 dem zufolge die eiuzelnen Auhangsorgane des embryonalen Darmes, wie Lunge, 
 Leber, &c., Ausstiilpungeu des Darmkanals w.aren. Vielmehr sieht man sich 
 genothigt, an diesen sJimmtlichen Organen nur insoferne das Darmdriisenblatt 
 als betheiligt anzusehen, als diess bloss Epithelialgebilde, fiir die Auskleidung 
 der einzelnen Ausfilhrungsgange oder deren Verzweigungen abgiebt. Die Ele- 
 mente, welche das Parenchym dieser Organe ausmachen, werden zumeist der 
 Urwirbelmasse entnommen, welche den Darmkanal lungiebt," pp. 88 — 90. Al- 
 though Professor Schenk here speaks of the doctrine to which I have referred in 
 the text as being the view taken by "the older authors," it is still supported by 
 writers who are decidedly modem. Professor Boll has lately published a 
 pamphlet (Das Princip des Wachsthums, Berlin, 1876) which aspires to supply 
 the long-delayed reason why, in the development of glands, the mucous mcmbrauo 
 grows inwards as Eemak described. The formation of the gland, he concludes, is 
 not the result simply of the epithelial membrane growing inwards, but it is the 
 issue of a conflict between the epithelial membrane growing iu-s'ards and the blood- 
 vessels growing outwards. 
 
 2 Anatom. and rath. Observations, 1815, p. 32; or Anat. Memoirs, Vol. ii. 
 p. 125. 
 
Ch. IV.] OF THE BREAST. 87 
 
 " The blastema, which announces the approaching formation of 
 a gland in the embryo, in some instances precedes, and is in other 
 instances contemporaneous with, the conical blind protrusion of the 
 membrane upon the surface of which the future gland is to pour 
 its secretion. 
 
 " In certain instances it has been observed that the smaller 
 branches of the duct are not formed by continued protrusion of 
 the original blind sac, but are hollowed out independently in the 
 substance of the blastema, and subsequently communicate with the 
 ducts. 
 
 " It appears to be highly probable, therefore, that a gland is 
 originally a mass of nucleated cells, the j)rogeny of one or more 
 parent cells; that the membrane in connexion with the embryo 
 gland may or may not, according to the case, send a portion of the 
 membrane, in the form of a hollow cone, into the mass; but whe- 
 ther this happens or not, the extremities of the ducts are formed 
 as closed vesicles, and then nucleated cells are formed within 
 them, and are the parents of the epithelium cells of the j^erfect 
 organ." 
 
 These summar}^ statements are prefaced by the remark : 
 
 " We require renewed observations on the original develop- 
 ment of glands in the embryo. From the information we possess, 
 however, it appears that the process is identical in its nature with 
 the growth of a gland during its state of functional activity." 
 
 It is the latter subject that chiefly engaged Goodsir's attention. 
 His observations were made on animals in which the glands have 
 particular periods of functional activity. The sexual glands of cer- 
 tain vertebrates as well as invertebrates were studied during those 
 periodical processes of "development, maturity and atrophy," which 
 they undergo from season to season ; and it is with the periodical 
 processes that he says the embryonic development is identical. The 
 great significance of that remark will be discussed in the next 
 chapter ; but a short extract will be useful at present to illustrate 
 further Goodsir's view of the embryonic development of acini. 
 The above-quoted statement that " the extremities of the ducts 
 are formed as closed vesicles " will become clearer from the follow- 
 ing, relating to the periodical " develojDment :" 
 
 " An acinus is at first a single nucleated cell. From the nu- 
 cleus of this cell others are produced. From these again, others 
 
88 THE DEVELOPMENT [Pt. I. 
 
 arise in the same manner. The parent cell, however, does not dis- 
 solve away, but remains as a covering to the whole mass, and is 
 appended to the extremity of the duct. Its cavity, therefore, as a 
 consequence of its mode of development, has no communication 
 with the duct. 
 
 "The original parent cell now begins to dissolve away, or to 
 burst into the duct at a period when its contents have attained 
 their full maturity. This period varies in different glands, 
 according to a law or laws peculiar to each of them. 
 
 "In the gland, there are a number of points from which acini 
 are developed, as from so many centres. These I name the 
 germinal spots of the gland." 
 
 The following account of the development of the mamma in 
 the guinea-pig will be found to sujjport Goodsir's general view of 
 gland-development\ 
 
 In the new-born guinea-pig the mamma is found in the 
 inguinal region on each side, extending obliquely downwards and 
 backwards from the single nipple. If a vertical section be made 
 in a line running from the nipple inwards toward the symphysis 
 pubis, the tract of mammary tissue will be exposed. In a prepara- 
 tion hardened in alcohol it appears as a somewhat narrow strip of 
 ■whitish tissue about half an inch in length and running through 
 the midst of a large body of fat-tissue. From the long central 
 tract, short offshoots project into the surrounding fat. 
 
 In embryos about two-thirds grown, the mammary rudiment 
 is found to have the same relative position and extent. In the 
 section of the prepared tissue there is not the same naked-eye dis- 
 tinction between the tract of mammary tissue and the surround- 
 ing fat. Mamma and fat together form a small pear-shaped body 
 beneath the skin, the pointed end terminating anteriorly at the 
 nipple. 
 
 At both of the stages here mentioned, the developing mamma 
 is found to have relatively its full dimensions, and to lie amidst a 
 
 1 The similarity of methods and conchisions with those of Goodsii-, both in 
 this and former chapters, is qiiite undesigned, for I had published the investigations 
 on the periodical processes of the breast and a preliminary notice of its develop- 
 ment, before I became acquainted with his memoir on "Secreting Structiu-es." 
 That paper does not specially refer to the breast, but its general conclusions 
 agree closely with thosp that I q,rrive at in the particular case. 
 
Cii. IV.] OF THE BREAST. 89 
 
 quantity of fat-tissue ; the plan of the organ is, as it were, hiid 
 down over its whole extent. In the earlier of the two stages, the 
 acini have not begun to a}3pcar distinctly, but only the system of 
 ducts. At the time of birth, however, the acini are found in active 
 development, and within a few days after birth they are sometimes 
 found to have so great an exjaansion as to completely overshadow 
 the system of ducts. 
 
 The development of the system of ducts can be best considered 
 after that of the acini has been described, for reasons that will 
 appear. The contention of the first part of this chapter is that the 
 ^cini, or essential secreting structure of the breast, develope from a 
 matrix-tissue at numerous scattered points or centres ("germinal 
 spots" of Goodsir) ; that the matrix-tissue or embryonic cells are 
 the same from which the fat surrounding the mamma developes ; 
 and that the mode of development of the acini is, for the indi- 
 vidual cell, exactly the same process as in the development of the 
 fat-lobules. The mammary gland would therefore be a further 
 specialisation of fat-tissue, and a product of the mesoblast. It has 
 been said that the opinion of the mamma being a complex exten- 
 sion of the ectoderm has gained currency somewhat too easily, and 
 that a sufficient proof of this process of continuous extension from 
 the skin would be really an arduous matter. But if it is difficult to 
 prove that origin of the mamma by direct evidence, it is no less 
 difficult directly to disprove it. To make out a parallelism, how- 
 ever, between the development of the mammary acini and the 
 development of the adjacent lobules of fat is perfectly fresh 
 ground ; and that parallelism, and the identity of the matrix iu 
 the two cases, will be the issue raised. 
 
 The resemblance between the mode of origin of the fat-lobules 
 and of the mammary acini is first suggested by that stage of the de- 
 velopment of the breast which is represented in Fig. 1 (of the plate). 
 The figure represents a mammary lobule in a guinea-pig four days 
 old. The exjmnsion of the lobule is very considerable, so much so 
 that the branches of the duct are completely obscured by the 
 acini. Each acinus is a rounded space distended by large vacuo- 
 lated cells like fat-cells, among which a number of nuclei lie at 
 various depths. If reference be now made to Fig. 2, which shows 
 the condition of the breast in a guinea-pig that was estimated 
 from its size to be about two or three weeks earlier, some idea 
 
90 THE DEVELOPMENT [Ft. I. 
 
 will be obtained of the way in which this great expansion of 
 secretory structure comes about. In the earlier of the two stages 
 of development, there are present only the ducts, representing the 
 framework of the gland, and around them a quantity of embryonic 
 tissue. It is from this field of embryonic tissue that the cluster 
 of acini, as shown in Fig. 1, have sprung up. Now, at the earlier 
 stage of development shown in Fig. 2, a development of the 
 matrix-cells into fat-cells has already taken place at various 
 points; the space immediately surrounding the rudimentary lobule 
 in the figure is occupied by fat-tissue, although the drawing has 
 not been extended so far. It will thus appear, that if the mam- 
 mary acini and the fat-lobules that surround them develope from 
 the same matrix, the fat-development is considerably in advance 
 of the development of the secreting structures. The contention is 
 that the embryonic tissue nearest to the already formed ducts is 
 left to undergo a later and somewhat more special differentiation. 
 
 It might therefore be supposed that the mammary region of the 
 guinea-jjig contained within itself, at various periods of embryonic 
 life, all the elements for a comparison between the development 
 of collections of fat-cells and the development of acini. The 
 assertion is that from one and the same matrix there develope first 
 a certain number of fat-lobules, and afterwards the acini. The 
 peculiar cellular transformation that leads to the production of 
 fat-lobules in certain parts of the matrix is the same that, two or 
 three weeks later, leads to the formation of acini in the survivinsf 
 or untransformed part of the matrix. But that apparently simple 
 question cannot be put to a satisfactory proof in the guinea-pig 
 alone. It so happens that the parallelism between the fat- 
 development and the development of acini can be much more 
 clearly traced by studying the course of fat-development in the 
 inguinal region of the foetal kitten ; and the reason of this will 
 now be explained by a statement of the anatomical characters of 
 the mamma in the two animals. 
 
 In the adult guinea-pig the mamma on each side is a flattened 
 pyramidal body, the apex of which is at the nipple and the base 
 towards the gi'oin ; its length is nearly two inches. The mammary 
 structure proper is of a reddish-brown colour, and is surrounded on 
 all sides by a large quantity of fat. In a vertical section of the 
 entire organ, including the fat, the brownish parenchyma of the 
 
Cn. IV.] OF THE BREAST. 91 
 
 gland appears as a central wedge, sending out processes into and 
 interlocking witli the fat on all sides. The mamma is not separa- 
 ble from the fat without dissection, but the mamma and fat 
 together form a flattened pyramidal mass which is loosely connected 
 with the surrounding tissues. On the other hand, the mamma 
 of the cat on each side is an extensive broad and flattened strip of 
 glandular substance extending from the groin to the upper part of 
 the thorax, and supplied with four or five teats at intervals corre- 
 sponding to the systems of ducts of four or five originally separate 
 glands. It is only at the inguinal end that there is a considerable 
 quantity of fat ; it there forms a thick cushion beneath the gland. 
 The mamma is an expanded layer or stratum of purely glandular 
 structure, which rests, chiefly at its inguinal end, upon a cushion 
 of fat. Whereas in the guinea-pig the fat could be separated from 
 the gland only by a thorough dissection, in the cat it could be 
 at once removed in a separate layer. 
 
 Corresponding to those anatomical differences in the adult con- 
 dition of the two animals, there are remarkable differences in the 
 development. What may be called the mammary fat developes in 
 the kitten not only in the plane beneath the mamma, but it 
 developes as an independent organ. This organ is &,fat-hody, not 
 differing, except in its ulterior development, from the fat-bodies 
 that occur in other classes of animals in various situations and 
 endowed with various functions. But there is another important 
 difference between the mammary fat of the kitten and that of the 
 guinea-pig. In the foetal guinea-pig the fat has passed through 
 its embryonic transformation and has reached its nearly perfect or 
 mature form two or three weeks before birth. In the kitten the 
 fat, developing as a fat-body, is still undergoing its developmental 
 changes up to, and in some cases a few days beyond, the time of 
 birth. The development of the mammary fat is a very much 
 tardier or more gradual process in the kitten than in the guinea- 
 pig. From this circumstance, and from the circumstance of its 
 development as a separate organ or fat-body, or rather from both 
 circumstances combined, the comparison can be best made between 
 the mammary fat of the kitten, on the one hand, and the acini of 
 the guinea-pig on the other. An interval of two or three weeks 
 occurs in the guinea-pig between the fat-development and the for- 
 mation of the acini, and the same interval of time occurs between 
 
92 THE DEVELOPMENT [Pt. I. 
 
 the completion of the fot-developraent in the guinea-pig and that 
 in the kitten ; so that, although the mammary acini in the new- 
 born guinea-pig are comparable to its fat-lobules two or three 
 weeks before birth, they are also comparable to the fat-lobules of 
 the kitten of its own age ; and that comparison is much the more 
 striking of the two. There is reason to think that the develop- 
 ment of the mammary acini in the kitten follows that of the fat- 
 lobules at a short interval, as in the guinea-pig, and that it 
 takes place on the upper surface, or adjoining the youngest part 
 of the fat-body; but the development of the acini will be explained 
 fur the guinea-pig only. The kitten will therefore be used to ex- 
 jalaiu the fat-development in the inguinal region, and the guinea- 
 pig to explain the mammary development ; and it will be most 
 convenient, in order to make a comparison between the two pro- 
 cesses, to commence with an account of the former. 
 
 The description now to be given of the development of the fat- 
 tissue in the inguinal region of the kitten, or, in other words, of the 
 inguinal fat-body of the kitten, does not differ essentially from that 
 given by Toldt\ The chief points of immediate interest stated by 
 that author are as follows. In the earlier months of embryonic 
 life the fat-cell is a round and finely granular cell with a nucleus, 
 but without a cell-wall, and differing from a colourless blood- 
 corpuscle only in point of size ; at this stage the fat-cell is already 
 differentiated from a connective-tissue cell. The development of 
 the fat-tissue is always directly bound up with the existence of a 
 system of blood-vessels ; if three or four cells are found grouped 
 together as the rudiment of a fat-lobule, this grouping is already 
 in connexion with blood-vessels. By reason of the characteristic 
 and independent system of blood-vessels for the fat-tissue, there is 
 suggested a close analogy between the aggregate of fat-lobules and 
 an acinous glands Those aggregates of fat-lobules are the fat- 
 bodies. It is in the form of independent organs that the future 
 fat-tissue occurs in the embryo stage of some mammalians, while 
 these bodies persist through life in batrachians. In insects, they 
 are organs of great importance, and of various functions. The fat- 
 
 1 Sitzungsheriehte der Wiener Akademie, 1870, p. 451. 
 
 ^ Professor Toldt's words are: "There can be no more perfect analogy than 
 that which subsists between the fat-tissue and the so-called acinous glands, in 
 respect of the distribution of their blood-vessels." 
 
Ch. TV.] OF THE BliEAST. 93 
 
 body in the inguinal region of the new-born kitten is one of those 
 described by Toldt ; one of the figures in illustration of his paper 
 gives an excellent idea of the perfect and independent vascular 
 supply of the fat-body. 
 
 Tiie pair of fat-bodies in the inguinal region of the new-born 
 kitten are at first sight exactly like a pair of secreting glands. 
 They are about half an inch in lengthy somewhat curved, and 
 placed back to back ; their posterior end is thick and rounded, and 
 anteriorly they become narrower and pointed. Their anterior 
 ends, at the time of birth, correspond almost exactly to the inguinal 
 pair of teats of the kitten. The blood-vessels may be seen ramify- 
 ing on their lower and flattened surface, a large trunk running 
 exactly in the centre from behind forwards, and giving off branches 
 symmetrically on each side. The bodies have a loose fibrous in- 
 vestment, which isolates them throughout most of their extent ; 
 but at the upper and anterior end they cannot be detached 
 without some tearing of their substance. In other words, their 
 anterior extremity is not definitely rounded and circumscribed as 
 the posterior is. They have the parenchymatous consistence and 
 pinkish colour of a secreting gland, when fresh ; after hardening 
 in spirit, they become more spongy. Microscopic sections of them 
 present a characteristic appearance. The lobules are large at the 
 posterior end of the body ; the smallest groups of cells are met 
 with at the upper and anterior end. Each lobule is a closely 
 packed mass of large cells with very striking characters (Fig. 3, C). 
 The substance of the cells is a granular protoplasm, and in all of 
 them it is occupied by one or more vacuoles. Generally there is 
 only one large vacuole occupying the lower segment of the sphere, 
 while the protoplasm is chiefly massed at the other pole of the 
 cell, and appears in section as a crescentic band. The nucleus has 
 usually an excentric position in the midst of the crescent-shaped 
 band of cell-substance. The boundary of the cell, or the wall of 
 the vacuole, is also formed by the same granular substance, form- 
 ing a narrow belt or ring at the opposite pole. These cells have 
 all the character of secreting cells ; if there were an outlet for 
 their fluid contents, the fat-body might be viewed as a gland. 
 But a few days after birth, this developmental appearance passes 
 away ; the cells of the fat-body are so completely excavated and 
 filled with fluid that their granular substance disappears, and the 
 
94 rriE DEVELOPMENT [Pt. I. 
 
 original protoplasmic cell is represented only by a thin sac-like 
 membrane filled with an oily fluid. Instead of the vacuolated cell 
 giving up its fluid contents, as in a secretory process, it retains the 
 fluid permanently. At the same time the individual cells expand 
 greatly, and the fat-body becomes a mass of fat which extends 
 forwards and laterally. 
 
 Many of the cells found within the developing acini of the 
 mamma (in the guinea-pig) have the closest resemblance to the 
 vacuolated protoplasmic cells of the fat-body (in the kitten), and 
 this resemblance is so striking, that the young fat-lobules might 
 be mistaken for portions of the actual gland. There is, however, 
 no identity between the fat-body, or any part of it, and the mam- 
 mary secreting structure ; but there is an identity in the matrix- 
 tissue of both, and a remarkable parallelism in their respective 
 developments. This parallelism will now be explained, first by 
 tracing the fat-body (of the kitten) backwards to its origin, and 
 afterwards by describing the mode of origin of the mammary acini 
 (in the guinea-pig). 
 
 The various representative stages of the fat-body are shown 
 in A, B, and C of Fig. 3, that shown in C being the stage just 
 described above as immediately preceding the final or mature con- 
 dition of the fat-tissue. The condition in B, taken from an embryo 
 kitten about 8|- inches long, or about two-thirds grown, is the 
 characteristic earlier stage ; while A, from an embryo kitten 2^ 
 inches long, shows the earliest collection of cells that can be identi- 
 fied as going to form a fat-lobule. The fat-body, which, at the acme 
 of its development, is an isolated parenchymatous organ with a sepa- 
 rate vascular supply, commences at many scattered points in the 
 inguinal region. The small groups of cells that afterwards become 
 fat-lobules first show themselves at the extreme inguinal end of 
 the abdominal wall. Between the abdominal muscles on the lower 
 side and the rete inucosum on the upper, a wedge-shaped mass of 
 gelatinous tissue occurs, the pointed end being forwards. It is in 
 this subcutaneous matrix that the clusters of fat-cells take origin. 
 The appearance of the matrix-tissue is shown in Fig. 8, A. Spindle- 
 cells occur somewhat sparsely in the midst of a ground-substance 
 which is hyaline when fresh, but assumes a granular or thready 
 appearance after coagulation by reagents. At an earlier joeriod, 
 the embryonic cells of this region are relatively much more nu- 
 
Cn. IV.] OF THE BREAST. 95 
 
 meroiis. The spindle-cells arc the ordinary embryonic cells of 
 the mesoblast, persisting to a comparatively late period of foetal 
 life. At a number of points, such as the one shown in the centre 
 of the figure, cells are massed more closely. There is no reason 
 to suppose that these groups of cells are derived from any other 
 source than the spindle-cells of the part. The group begins to 
 form by the subdivision of one of the embryonic cells ; the 
 progeny are more globular, and with a larger amount of proto- 
 plasm round the nucleus. The differentiation towards a fat-lobule 
 has already begun, and the divergence from the spindle-cell or cell 
 of the connective-tissue type goes on increasing. The small 
 clusters of spherical protoplasmic cells expand ; the cells at the 
 same time become larger and more like epithelium, the growing 
 lobule having the appearance shown in Fig. 3, B. In the oldest 
 part of the fat-region, the lobules are already large and packed 
 closely together ; although the fat-body is not yet encapsuled or 
 isolable as a whole, yet the space between the skin and the 
 abdominal muscles is readily seen to be occupied by a special 
 kind of tissue, forming a wedge-shaped mass which, in alcohol pre- 
 parations, has a whitish appearance. The largest lobules are many 
 times the size of the section figured, and their cells, representing 
 a passing phase in the development, often have a close resemblance 
 to cubical or polyhedric epithelial cells. About the time of birth 
 the fat-body has acquired the appearance of being an independent 
 organ, and the cells are all vacuolated as in Fig. 3, C. During the 
 development of the fat-body, one or more lymphatic glands have 
 developed towards its lower surface and as if enclosed by it. 
 
 Throughout the growth of this body or organ, the development, 
 as has been already stated, is not in an equally forward state at all 
 points. The groups of fat-cells first appear at the extreme inguinal 
 end of the region, and the order of appearance of the various 
 centres of development is, throughout the whole process, from 
 behind forwards and from the deeper parts towards the surface. 
 Again, when the fat-bodies are already of considerable size in the 
 groin, the corresponding formation of fat- tissue at separate points, 
 beneath the three or four anterior pairs of teats in the kitten, has 
 not advanced beyond the first stage of small scattered groups of 
 embryonic cells. In these anterior regions of the abdominal and 
 thoracic walls, the fat is at no period of its existence found as a 
 
96 THE DEVELOPMENT [Pt. I. 
 
 separate body or organ, althougli the early stages of its formation 
 are the same as in the actual fat-body of the groin. Limiting the 
 attention again to the latter, the first steps of development may be 
 observed at its anterior end, and on its upper surface, and even in 
 the tracts of embryonic tissue that lie between the lobules, at a 
 time when the development as a whole is well advanced. It is 
 beside those tardier points, at the anterior end and on the upper 
 surface, that the first appearance of mammary secreting structure 
 must be looked for. In the cat I have not succeeded and indeed 
 have hardly attempted to trace the formation of the mammary 
 acini from the embryonic tissue of the upper surface of the fat- 
 body. If it be necessary to carry out the alleged parallelism in 
 the same animal, it must suffice to state the hypothesis that in the 
 cat the mammary acini begin where the fat ends, and that the 
 latest differentiation of the embryonic cells in the fat-region is 
 appropriated to a still more particular purpose, that is, to become 
 the secreting stnicture of a gland. But at this point the field of 
 observation must be changed from the cat to the guinea-pig. 
 
 It has been already explained that, whereas in the cat the 
 lower end of the mammary gland lies altogether above the fat- 
 body, or rather above the cushion of fat into which the latter is 
 changed ; in the guinea-pig, on the other hand, the mamma and 
 the fat-tissue are interlocked on all sides, and the mamma is 
 limited to the inguinal region. Whatever these differences may 
 signify, the guinea-pig is on account of that characteristic a more 
 convenient animal on which to demonstrate the present theory of 
 mammary development. The fat-lobules of the mammary region 
 in the guinea-pig do not form a fat-body ; they never pass through 
 the remarkable gland-like phase that the fat-body of the kitten is 
 so well adapted to illustrate. But, on the other hand, the mam- 
 mary development takes place in the midst of the fat-development, 
 and it is therefore easier to show that the matrix-tissue of the two 
 kinds of structure is the same. The mere contiguity does not 
 necessarily afford any proof, but it greatly simplifies the state- 
 ment of the problem and the ^presentation of evidence. For the 
 same reason as in the guinea-pig, the udder of the fo3tal calf would 
 probably be found very useful for putting this question to the 
 proof. In a foetal cow-calf, about half-grown, the udder is found 
 to have its proper shape and its relative size, being a conical 
 
Ch. IV.] OF THE BREAST. 97 
 
 eminence completely raised above the skin, upwards of an inch in 
 diameter at its base, and somewhat less in height. Two pairs of 
 teats are also perfectly developed, and the rudiments of large ducts 
 or lacunfB are found over a small space directly beneath the teats. 
 The substance of the udder is a gelatinous mass, in the midst of 
 which lie a large number of small round bodies of a somewhat 
 different texture, like tubercles in the substance of an organ. These 
 round tubercle-like bodies are distributed throughout the whole 
 substance of the udder quite regularly, and it might be supposed 
 that they were the already developed lobules of the gland, visible 
 through the gelatinous stroma. On examination, however, they 
 are found to be collections of fat-cells, developing from the gela- 
 tinous embryonic tissue of the udder at many regularly distributed 
 centres. The abundant gelatinous tissue in which they occur is of 
 the embryonic kind usual in the mesoblast, containing numerous 
 spindle-cells and a large amount of hyaline intercellular substance. 
 The hypothesis, as regards this form of udder, would be that the 
 acini develope throughout the whole udder from the remaining 
 embryonic tissue, in much the same way, and from a correspond- 
 ing number of independent centres, as the fat-lobules had done 
 before them. It was not possible, however, to put the hypothesis 
 to the test, from want of suitable material. 
 
 In describing the development of the mamma in the guinea- 
 pig, it is convenient at first to assume the development of the 
 ducts, and to consider the acini apart. In Fig. 2 is shown a portion 
 of a duct and terminal branches. The tissue surrounding it is a 
 part of the embryonic tissue that the whole wedge-shaped body in 
 the gi'oin at one time consisted of. The ducts now penetrate the 
 centre of it, and the fat-tissue lies immediately outside it. Those 
 surviving tracts of embryonic tissue are seen to contain groups of 
 cells at various points, which, at first sight, resemble the aggre- 
 gates of embryonic cells that the fat-lobule had developed from. 
 Turning: now to Fiof. 1, showincr the state of the mamma two or 
 three weeks later, the place of the embryonic tissue round the duct 
 is found to be taken by a large collection of acini, completely over- 
 shadowing the central duct to which they are related. The rudi- 
 mentaiy or potential lobule of the earlier period has expanded 
 into a voluminous group of acini, each of which resembles rather a 
 small collection of fat-cells than the epithelium-covered recess of a 
 c. c. 7 
 
98 THE DEVELOPMENT [Pt. I. 
 
 secreting gland. The acini are essentially collections of fat-cells 
 which have developed from the remaining tracts of the matrix- 
 tissue by the side of the ducts, and which have undergone exactly 
 the same process of transformation in the individual cell that the 
 fat-tissue had undergone at an earlier period. The original wedge 
 of embryonic tissue in the inguinal region has become fat-tissue, 
 except along certain central tracts or planes in which the already- 
 formed ducts lie ; the ducts, and the embryonic tissue among 
 which they lie, form, as it were, an island of the original matrix 
 that has escaped becoming fat-tissue. At a later stage of develop- 
 ment, the encroachment of the fat is still greater, and the 
 embryonic tissue remains only as narrow strips along the sides of 
 the ducts and round their terminations. It is from these narrow 
 margins of the original matrix that the acini develope. The great 
 expansion of acini shown in Fig. 1 implies a very considerable 
 development from so narrow a basis; but the process of expan- 
 sion becomes intelligible by observing the mode of development in 
 a single acinus. It is here that the observations now to be recorded 
 agree with the general conclusions on gland-development arrived 
 at by Good sir. 
 
 If, in new-born guinea-pigs, the margins of embryonic tissue 
 which run along the sides of the ducts, or extend from the ends of 
 them in narrow planes through the fat, be examined closely, there 
 will be frequently seen the appearances drawn in Fig. 4, A^ B, 
 and C, under a magnifying power of about 600 diameters. There 
 is first distinguished ^ small cluster of cells, as in ^; one of the 
 cells is vacuolated and has a large nucleus on the periphery, two 
 other cells have a conical extension of protoplasm, and there is on 
 the other side an elongated nuclear cell. In the margin of tissue 
 by the side of the duct, these four cells are a distinct and isolated 
 cluster. This margin of tissue can be seen, by careful examination, 
 to contain many such, and among them groups of the kind shown 
 in B. This group is more obviously an independent cluster or 
 colony ; the outer row of cells have an elongated crescentic form, 
 and some of them have processes. which appear to complete the 
 circuit and retain the whole within a definite boundary. In C a 
 similar appearance is shown ; the cellular development is found in 
 the midst of one of the narrow strips of embryonic tissue that 
 extend from the duct through the fat. At the lowest part of the 
 
Ch. IV.] OF THE BREAST. 99 
 
 figure is a cluster of very minute nuclear cells ; above them are 
 two other rudimentary acini with better marked characters. The 
 lower of the two contains two large vacuolated cells, not distinguish- 
 able from fat-cells ; the resemblance to fat-cells is the same that 
 occurs in the later stage of development (Fig. 1), and the groups 
 of vacuolated cells are as clearly a part of the mammary structure 
 in the one case as in the othei'. It appears probable that the 
 regular spherical arrangement of the cells that form an acinus is 
 produced by means of the extensive vacuolation of one or more of 
 the cells. The periphery of the vacuolated cells represents the 
 circular boundary of the acinus. If the figures B and G are com- 
 pared, the crescentic cells in B that form the boundary of the 
 acinus will be found to correspond to the peripheral nuclei of the 
 larsfe vacuolated cells in G. All the cells of the cluster, more or 
 less, undergo the vacuolation at one time or another ; the earliest 
 vacuolations seem to determine the form and expansion of the 
 acinus, while in the acini of Fig. 1 the process is seen to be quite 
 general throughout the large number of cells that the acinus now 
 contains. These appearances, apart from their bearing on the homo- 
 logy with the fat-lobules, seem to point to a less rigid law of 
 acinous development than that laid down by Goodsir. According 
 to Goodsir, " an acinus is at first a single nucleated cell. From 
 the nucleus of this cell others are produced. From these, again, 
 others arise in the same manner. The parent cell, however, does 
 not dissolve away, but remains as a covering to the whole mass, 
 and is appended to the extremity of the duct. Its cavity, there- 
 fore, as a consequence of its mode of development, has no commu- 
 nication with the duct." According to this view, based chiefly on 
 studies of invertebrate or lower vertebrate animals, the boundary 
 of the acinus is formed by the distended wall of a single cell, in 
 the parent cavity of which all the cells of the acinus, or the future 
 epithelium, have developed. However that may be; the present 
 description agrees with that of Goodsir in the important point that 
 the development of acini is not by means of protrusions of the 
 ducts so as to form infundibula or recesses at many points along* 
 their course, but that it is an interstitial development from the 
 embryonic tissue that surrounds the ducts. Those margins of 
 embryonic tissue, it has been said, become extremely narrow by 
 the time that the system of ducts and the surrounding fat are 
 
 7—2 
 
100 THE DEVELOPMENT [Ft. I. 
 
 completely developed ; so much so, that at some places they might 
 be maintained to be nothing more, in relation to the ducts, than a 
 submucous row of cells. In that point also there is an agreement 
 with the observations of Goodsir, and it will be seen from the 
 quotation that he did not find the circumstance just mentioned to 
 be inconsistent with the notion of interstitial development. The 
 observation of Goodsir does not refer to the embryonic develop- 
 ment, but to those periodical " developments" of glands from season 
 to season, which he compared to their embryonic development. 
 The example chosen is the testicle of a shark {Squalus cornuhicus) 
 in its various stages as it advanced from the resting state to the 
 active secretory state at the time of sexual vigour. The " develop- 
 ment" of the acini had the following stages : 
 
 " 1st. A single nucleated cell attached to the side of the duct, 
 and protruding, as it were, its outer membrane. 
 
 " 2nd. A cell containing a few young cells grouped in a mass 
 within it; the parent cell presenting itself more prominently on 
 the side of the duct. 
 
 " 3rd. A cell attached by a pedicle to the duct, the pedicle 
 being tubular and communicating with the duct ; the cell itself 
 being pyriform, but closed and full of nucleated cells. 
 
 " 4th. Cells larger than the last, assuming more of a globular 
 form, still closed, full of nucleated cells, and situated more towards 
 the surface of the lobe. 
 
 " 5th. The full-sized vesicles already described as situated at 
 the surface of the lobe. These vesicles are spherical, perfectly 
 closed ; that part of the wall of each which is attached to the 
 hollow pedicle forms a diaphragm across the passage, so that the 
 vesicle has no communication with the ducts of the gland." When 
 the contents of the acini are mature, the diaphragm, says Goodsir, 
 is broken through, and the contents of the acini are discharged into 
 the ducts. 
 
 Having described in the guinea-pig the mode of origin of the 
 acini by the side of the ducts, and their appearance in a fully- 
 expanded lobule at the time of birth, it will now be possible to 
 trace the parallelism with the fat-development at all points. The 
 cluster of four cells iu Fig. 4, A (mamma of guinea-pig), is in every 
 respect the same as those small groups of cells that indicate the 
 beginning of a fat-lobule iu the extreme inguinal region of the 
 
Cn. lY.] OF THE BREAST. 101 
 
 foetal kitten ; the presence of one or more vacuolated cells, around 
 which the others are grouped, is no less characteristic of the latter 
 than of the former, and this initial similarity is as striking as the 
 resemblance at more advanced periods. In the case of the kitten, 
 these " germinal spots " of the matrix proceed to form the lobules 
 of a fat-body, which has many points of resemblance to a gland ; in 
 the case of the guinea-pig they arise by the side of a system of 
 ducts already there, and instead of forming a gland-like fat-body, 
 they become the serviceable acini of a racemose gland. The stages 
 A, B, and C of Fig. 3 (fat-lobule of kitten) may be compared to 
 A, B, and C of Fig. 4 (mammary acinus of guinea-pig) ; while the 
 expanded mammary lobule of Fig. 1 may be compared to a cluster 
 of fat-lobules of the same gland, as they existed three weeks 
 earlier. There is thus an identity as regards the matrix-tissue, 
 a similarity in the initial grouping of the embryonic cells, and a 
 similarity in the destiny or subsequent transformations of the cells, 
 that can hardly fail to arrest the attention. That destiny is that 
 they become spherical, assuming a protoplasmic investment round 
 the nucleus, and that they undergo a process of vacuolation. The 
 two distinguishing features of the groups of cells that become the 
 mammary acini may be said to be, (1) that they are the last of the 
 embryonic cells to undergo their peculiar development, and (2) 
 that they are ranged along the sides of ducts. In those two cir- 
 cumstances lies the differentiation, in the individual development, 
 of the mammary acini from the other parts of the fat-body or of 
 the inguinal fat-region. The comparative anatomy of the breast 
 will throw some light on the way in which that differentiation has 
 originated. It will be convenient to introduce this subject at once, 
 in treating of the next section of the inquiry, viz. the development 
 of the ducts. 
 
 In the guinea-pig a system of ducts is found extending through- 
 out the entire region of the future gland at least two weeks before 
 any acini can be distinguished as such, and both in the foetal 
 kitten and guinea-pig the nipples are well formed at a still earlier 
 period. This order of development in the individual rodent or 
 carnivorous animal is a remarkable reversal of the order of appear- 
 ance of the various structures in the phylogenetic succession of 
 mammalian animals. In the Ornithorhynclius and Echidna there 
 is no nipple, and there are, strictly speaking, no ducts. The 
 
102 THE DEVELOPMENT [Pt. L 
 
 mamma is composed of about 100 — 200 flask-shaped follicles, 
 each of which terminates in the skin Ly an elongated neck. 
 The necks into which the follicles are drawn out, and by which 
 they are attached to the skin, may be regarded as an incipient 
 system of ducts. In each follicle there does not appear to be more 
 than a single central passage. The gland, in fact, belongs to the 
 follicular order of glands, which appear to be the primitive type of 
 secreting structures, and which often occur in the lower animals 
 where higher species have glands of the acinous type. The fol- 
 licles are generally elongated, the blind extremity being slightly 
 pointed. The substance of the follicle is a mass of secreting cells, 
 in whose midst a more or less regular passage exists near its 
 attached end, by means of which the secretion is poured out. It 
 has, indeed, been asserted by Owen^ that the structure of the 
 gland in the Echidna is "on the same general plan as that of the 
 mammary glands in higher mammals," and he states, in particular, 
 that the duct by which the follicle or lobule opens on the skin is 
 " directly continued from a canal which may be traced about half 
 way towards the fundus of the lobule ; the canal gives off numerous 
 short branches from its circumference, which subdivide and termi-- 
 nate in clusters of sub-spherical ' acini ' or secerning cellules." He 
 also states that the structure closely resembles that of the Orni- 
 thorhynclms. I have taken the opportunity of examining the 
 minute structure of the glands of both animals with reference to 
 their alleged acinous type. That of the Echidna was in its most 
 fully expanded state, while that of the Ornithorhynchus was in an ex- 
 tremely "involuted" condition, and had not one-tenth the volume of 
 the other. In former chapters I described at length the histological 
 nature of those periodical processes of " evolution " and "involution," 
 and among other noticeable points was this, that in the full develop- 
 ment of the secreting structure during lactation it was difficult to 
 see the system of ducts on account of the great expansion of the 
 acini which concealed them, while in the shrunken or involuted 
 state of the gland, when the function was in abeyance, the ducts 
 and clusters of terminal acini were everywhere jDrominent, and 
 their branching arrangement could be clearly traced. So that, if 
 the microscopic examination revealed no system of ducts and ter- 
 
 ^ Philosophical Transactions, 18G5. 
 
Cn. IV.] OF THE BREAST. 103 
 
 minal acini in the expanded gland of the Echidna, they ought at 
 least to have been quite visible in the involuted gland of the 
 Ornitho)-hi/nchus. But such was not the case. A single central 
 passage ran through each lobule, and it is probable that it had, 
 when fresh, a lining of columnar epithelium ; but a racemose 
 system of ducts and acini was certainly not present. The minute 
 structure of the mamma in those two animals will be briefly re- 
 ferred to afterwards ; for the present they may be considered with- 
 out doubt to be glands of the follicular type. The class of animals 
 that come next to the Monotremata, as regards the simplicity of 
 the breast-structure, are the Cetacea ; the marsupial animals in 
 this respect must be considered somewhat higher. In the porpoise 
 each ffland has a single common duct, which runs from a reservoir 
 under the nipple downwards to a considerable depth. Around this 
 central vertical duct the secreting follicles of the gland are grouped 
 in more or less horizontal planes at various depths. The structure 
 is still essentially follicular, but the follicles, instead of opening on 
 the skin each by its own duct, discharge into a common conduit at 
 various points'. It may be shown diagrammatically how the 
 racemose system of ducts could be produced by extending the 
 same principle of centralisation, of which the beginning is seen in 
 the cetacean mamma. With the racemose system of ducts comes 
 necessarily a subdivision of the parenchyma into acini. It is per- 
 haps impossible now to trace, in nature, all the steps by which this 
 modification came about, but it appears certain that the original 
 part of the mamma was its parenchyma, or the solid aggregates of 
 its secreting cells, and that the ducts, which greatly help to eco- 
 nomise the secretion, were acquired later. Now, as regards this 
 particular organ of the body, the principle that the embryonic 
 development in the individual is a brief recapitulation, in their 
 order, of all the acquisitions that have been made in the phylo- 
 genetic succession, does not hold good. The nipple is the latest 
 addition or improvement on the original mamma, but in the guinea- 
 pig and cat, and no doubt in animals generally, it is developed in 
 the skin long before the secreting structure is developed in the 
 lower stratum. The elaborate system of ducts constitutes the 
 
 ^ The resemblance of the cetacean mamma to that of the Oniithorhijnchns was 
 pointed out by Von Baer during the controversy that followed Meckel's discovery of 
 the gland in the monotreme. 
 
104 THE DEVELOPMENT [Pt. I. 
 
 other addition or improvement, but in the guinea-pig the plan 
 of them is laid down throughout the whole extent of the future 
 gland, before the essential secreting structure or parenchyma 
 makes its appearance as such. The early appearance of the 
 ducts has been duly noticed by the writers on the development 
 of the breast, and it is no doubt owing to the circumstance of the 
 ducts appearing first in the individual, growing as it were from 
 the rete mucosum downwards, that the gland has been taken to be 
 essentially a complex reduplication of the ectoderm, the recesses 
 or infundibula of which were utilised for secretion. It has been 
 pointed out, at the beginning of the chapter, that the theory had 
 been accepted without due consideration of the defective kind of 
 evidence on which it was based ; and in describing particularly the 
 development of the acini or the essential secreting part of the 
 gland, I have been led to an entirely different view of the develop- 
 ment of the whole organ. 
 
 But the question then remains, How comes it that this alleged 
 order of acquisition of the various parts of the mamma in the evo- 
 lution of animals, is reversed in the development of the organ in 
 an individual of the higher mammalians, such as the guinea-pig ? 
 In answering this question it will be necessary to make use of an 
 argument that is expounded in Mr Herbert Spencer's Principles 
 of Biology, and which appears to be an original generalisation of 
 that writer. The following extract^ will present the argument in 
 the briefest possible way. 
 
 " The same kind of structure is not always produced in the 
 same ways; and some allied groups of organisms have modes 
 of evolution which appear to be radically contrasted. The two 
 modes are broadly distinguishable as the direct and the indirect. 
 They may severally characterise the general course of evolution 
 as a whole, and the course of evolution in particular organs.... 
 Thus in the immense majority of articulate animals, metamor- 
 phoses, more or less marked and more or less numerous, are 
 passed through on the way to maturity. The familiar transfor- 
 mations of insects show us how circuitous is the route by which 
 the embryo-form arrives at the adult form among some divisions 
 of the Articidata. But there are other divisions, as the lower 
 Arachiida, in wliich the unfolding of the egg into the adult 
 
 ^ Principles of Biology, Vol. i. p. 371 et seq. 
 
Cn. IV.] OF THE BREAST. 105 
 
 takes place in the simplest manner: the substance grows to- 
 wards its appointed shape by the shortest route." 
 
 The general rule is that the indirect development character- 
 ises the most highly organised forms; conversely it is direct in 
 a large proportion of the lower types. One of Mr Spencer's 
 illustrations from social affairs is the followiuj; : 
 
 "A new town in the United States arises not at all after 
 the old method of gradual accumulations round a nucleus, and 
 successive small modifications of structure accompanying in- 
 crease of size; but it grows up over a large area according to a 
 predetermined plan; and there are developed, at the outset, 
 those various civil, ecclesiastical and industrial centres, which 
 the incipient city will require." 
 
 The argument then proceeds : 
 
 "Now on the hypothesis of evolution, all organs must have 
 been originally formed upon the indirect method, by the accumu- 
 lation of modifications upon modifications; and if the develop- 
 ment of the embryo repeats the development of ancestral races, 
 organs must be thus formed in the embryo. To a considerable 
 extent they are thus formed." 
 
 Mr Spencer then gives examples of organs that are formed 
 according to the indirect mode, and of other organs formed 
 according to the direct, and concludes that, on the hypothesis 
 of evolution, the direct mode of development in the latter class 
 must have been substituted for the indirect. 
 
 The bearing of this speculation on the development of the 
 mamma and its various parts, will now become apparent. As 
 there are certain classes of animals distinguished from other 
 classes by the "directness" of their development, and as there 
 are some organs so distinguished from other organs in the same 
 animal, so there are, in one and the same organ, some parts 
 developed "directly" and other parts developed "indirectly." 
 In the mammary gland, the nipple and ducts are developed 
 directly, and the secreting structure, or the acini, are developed 
 indirectly. In a mammary gland, the ducts were originally 
 spaces formed in the midst of the parenchyma by a modification 
 or adaptation of the same along certain central tracts. But the 
 developing gland in the individual (guinea-pig, for example) 
 does not repeat the particular steps of adaptation or modifica- 
 
106 THE DEVELOPMENT [Ft. I. 
 
 tion, inasmuch as the ducts arc laid down throughout their 
 "vvhole extent before the secreting structure appears. Tlie indi- 
 rect process of adaptation is replaced by a direct mode of origin. 
 It will now be suitable to introduce certain facts relating to the 
 origin of the ducts. 
 
 In the wedge-shaped body of embrj^onic tissue in the groin 
 of a foetal guinea-pig less than half grown, but with nipple 
 already formed, there are seen, extending backwards from the 
 nipple, certain narrow tracts of cells, which are simply the 
 embryonic cells of the part closely packed together. These 
 tracts of closely packed cells form at various points throughout 
 the embryonic mass independently of each other, and no con- 
 tinuous extension of them can be traced from the nipple. At 
 this period of development, the fat-lobules are beginning to 
 form from the same matrix, and it is difficult in some cases 
 to distinguish the elongated and narrow collections of rudimen- 
 tary fat-cells from the first mentioned linear aggregations of 
 cells. But there are some unambiguous cases in wdiich a dis- 
 tinct lumen can be seen in the narrow tract of cells, both in the 
 longitudinal and in the cross section, and it can hardly be 
 doubted that these are rudimentary ducts. The appearances 
 observed in the udder of the foetal calf, and in the mammary 
 region of the new-born rat, also favour this view of the origin of 
 the ducts. There is certainly no evidence of a process of 
 growth downwards from the rete mucosum under the nipple. 
 The mode in which a communication is formed between the 
 epidermic excavation or depression in the nipple, and the ducts 
 forming in the lower stratum, does not seem to present any 
 great difficulty of explanation. 
 
 The ducts are therefore held to be formed as aggregations of 
 the embryonic cells of the matrix along certain predetermined 
 lines; that preliminary point may be taken as established, and 
 it matters little for the present purpose -that no circumstantial 
 account can be given of the manner in which these cord-like 
 structures are formed into hollow tubes lined, in the mature 
 state, with columnar epithelium. Certain facts of a negative 
 kind can however be made out. In the course of formation of the 
 ducts, the embryonic cells do not appear to undergo any of the 
 very marked transformations that have been described for the 
 
Ch. IY.] of the breast. 107 
 
 cells of the aciui. As compared with those other structures that 
 the matrix-tissue gives origin to, the ducts may be said to be 
 formed out of the embryonic cells by the " direct " mode. In 
 Mr Spencer's words, " the substance grows towards its appointed 
 shape by the shortest route." It is this shortened or direct mode 
 of development that must be held to explain the early appearance 
 of the ducts. But the question still remains, how it is that in 
 the case of the ducts their development is by the direct, and 
 not by the indirect mode. This brings us back to Mr Spencer's 
 question : " How comes the direct mode of development to have 
 been substituted for the indirect ? " It will here be necessary to 
 resume the quotation from Mr Spencer. 
 
 " Supposing it were possible for a race of organisms to have 
 continued propagating itself through an indefinitely long period 
 without any change of conditions necessitating change of struc- 
 ture; there would be reached so complete a congi-uity between 
 the organic aggregate and its physiological units, that the units 
 would arrange themselves directly into a structure like that of 
 the adult organism : the germ would put on the proper charac- 
 ters of the species with little or no transposition of substance. 
 But in the absence of any such constancy of conditions and 
 structure, what may we expect? We may expect that where 
 the conditions and structure have been most constant, the mode 
 of development will be the most direct ; and that it will be the 
 most indirect w^here there have been the greatest and most 
 numerous changes in the habits and structures of ancestral 
 races of organisms... Between different parts in the same em- 
 bryo, there are unlikenesses in the method of formation, which 
 seem to have kindred meanings." 
 
 The former part of Mr Spencer's argument was made to 
 apply to different parts of the same organ, and it will now be 
 seen how the reasoning of the last quotation applies to the 
 ditferent modes of development within the mamma. 
 
 Comparing the parts of the mature mamma with one 
 another, the ducts are those parts that exist throughout the life 
 of the individual " without any change of conditions necessita- 
 ting change of structure;" whereas -the secretory activity of the 
 gland depends on constant changes in the cells of the acini. 
 This fact cannot be better illustrated than by reference to the 
 
108 THE DEVELOPMENT OF THE BREAST. [Pt. I. 
 
 periodical processes of " involution" and "evolution" described in 
 the first and second chapters. In those processes the ducts alone 
 continue as they were during the state of active lactation; the 
 whole processes of upfolding and unfolding take place in the 
 acini or true secreting parts of the gland. The ducts play a 
 purely mechanical part, and they continue without change from 
 the time that their development was completed. According to 
 Mr Spencer's argument as extended, it is under such circumstances 
 as these that the direct mode of development is substituted for the 
 indirect ; and it would appear to be in accordance with this prin- 
 ciple that the ducts of the mamma, though they are additions to 
 the primitive mamma and acquired comparatively late in the suc- 
 cession of mammalian animals, are in the individual higher mam- 
 malian developed from the matrix-tissue of the gland before the 
 •secreting structure itself. To borrow Mr Spencer's illustration of 
 a modern town of rapid growth, the ducts of the breast are like 
 the roads and streets that are laid down before any houses are 
 built. In the primitive state of a community, dwellings came 
 first and streets afterwards; in the primitive condition of the 
 breast, the secreting follicles of the gland existed first, and in 
 course of time they became connected together by a system of 
 ducts. 
 
 The conclusions of this inquiry are : 
 
 (1) That the mammary acini of the guinea-pig develope 
 at many separate points in a matrix-tissue; that the embryonic 
 €ells, from which they develope, are of the same kind that give 
 origin to the surrounding fat-tissue; and that the process of 
 development of the mammary acini is step for step the same 
 as that of the fat-lobules. 
 
 (2) That the ducts of the mamma develope from the same 
 matrix-tissue by direct aggregation of the embryonic cells along 
 predetermined lines ; that the ducts develope in the individual 
 guinea-pig before the acini, whereas in the phylogenetic suc- 
 cession the ducts are a later acquisition ; and that this reversal 
 of the order of acquisition of parts is in accordance with the 
 principle stated by Mr Herbert Spencer that, under certain cir- 
 cumstances, the direct mode of development tends to be substi- 
 tuted for the indirect. 
 
EXPLANATION OF PLATE. 
 
 Fig. \. Sectiou from the mamma of a guinea-pig four clays old. A 
 rounded or pyriform lobule made up of a large number of acini, the 
 cavities of which are distended with vacuolated cells. Nuclei are to be 
 seen among the vacuolated cells at various depths, x 300. 
 
 Fig. 2. Section from the mamma of a foetal guinea-pig about three 
 weeks before birth. System of ducts surrounded by embryonic tissue. 
 X 300. 
 
 Fig. 3. Showing the development of a fat-body. 
 
 A. From the fjelatinous tissue in the intjuinal region of a foetal 
 kitten 2^ inches long. In the middle of the figure, a small group of the 
 embryonic cells undergoing changes to become fat-cells, x 300, 
 
 B. From the glaad-like tissue in the same region of a fcetal kitten 
 3| inches long. Later stage of the fat-lobule, x 300. 
 
 C. From the fully-developed fat-body in the inguinal region of a 
 new-born kitten. A small lobule of the fat-body, showing the vacuo- 
 lated condition of the cells, x 300. 
 
 Fig, 4. From the mammaj of new-born guinea-pigs, x about 600. 
 
 A. Group of four cells in the embryonic tissue at the side of a duct. 
 The earliest appearance of an acinus. 
 
 B. Similar group showing a later stage of the acinus. 
 
 C. The upper part of the figure shows two acini developing in the 
 midst of a tract of embryonic tissue. Several of the cells extensively 
 vacuolated. 
 
Tis' 
 
 p:'\^f, 
 
 Lt^ ^o 
 
 '^ 
 
 
 \^'^y%-'^^ 
 
 
 
 riff i. 
 
 %f^'(j." . 
 
 
 ^2>' 
 
 
CHAPTER V. 
 
 THE DEVELOPMENT OF THE MAMMARY FUNCTION. 
 
 The development of function is one of those interesting de- 
 partments of physiology that the modern teachers of the science 
 do not profess to include in their progi^amme. It is remarked by 
 Haeckel that "the physiology of to-day does not concern itself 
 either with the functions of development or with the development 
 of functions. In its very one-sided tendency, it strives to bring 
 to the highest degree of completeness our knowledge of single 
 groups of functions — for example, the physiology of the organs of 
 sense, of muscle, of the circulation, &c. — while there are other 
 functions for which it has no concern." And he enumerates certain 
 functions of the organism that are not included in the scope of 
 physiology, as it is defined by the works of its professed teachers \ 
 I must leave such criticisms to those who can make them with due 
 authority, and, in a volume of mixed contents like the present, I 
 may treat of one of the neglected subjects in question, without 
 inquiring too curiously whether its interest is for physiologists or 
 for others. The consideration of most practical importance is that 
 one finds no classical writings on the development of function, and 
 it is therefore with some diffidence that I attempt, under an explicit 
 title, to treat of the development of the function of a particular 
 organ. 
 
 The breast is an organ on which the experiment may at least 
 be advantageously tried. Compared with an organ like the liver, 
 
 ^ Antliropogenie, p. 130. 
 
CHAPTER V. 
 
 THE DEVELOPMENT OF THE MAMMARY FUNCTION, 
 
 The development of function is one of those interesting de- 
 partments of physiology that the modern teachers of the science 
 do not profess to include in their programme. It is remarked by 
 Haeckel that "the physiology of to-day does not concern itself 
 either with the functions of development or with the development 
 of functions. In its very one-sided tendency, it strives to bring 
 to the highest degree of completeness our knowledge of single 
 groups of functions — for example, the physiology of the organs of 
 sense, of muscle, of the circulation, &c. — while there are other 
 functions for which it has no concern." And he enumerates certain 
 functions of the organism that are not included in the scope of 
 physiology, as it is defined by the works of its professed teachers \ 
 I must leave such criticisms to those who can make them with due 
 authority, and, in a volume of mixed contents like the present, I 
 may treat of one of the neglected subjects in question, without 
 inquiring too curiously whether its interest is for physiologists or 
 for others. The consideration of most practical importance is that 
 one finds no classical writings on the development of function, and 
 it is therefore with some diffidence that I attempt, under an exphcit 
 title, to treat of the development of the function of a particular 
 organ. 
 
 The breast is an organ on which the experiment may at least 
 be advantageously tried. Compared with an organ like the liver, 
 
 1 Anthropogenie, p. 130. 
 
110 THE DEVELOPMENT [Fr. I. 
 
 compared even with an organ like the lachrymal gland, it appears 
 late in the succession of animals. It may in fact be considered 
 to be the latest acquisition or addition to the glandular system 
 of animals. Moreover, in the class to which it is confined and 
 for which it supplies the class-name, it exhibits the more simple 
 type of gland-structure at one end of the scale, and the more 
 complex type at the other. The first part of this chapter will 
 treat of the development of the mammary function in so far as it 
 may be deduced from the character of the gland in the Monotre- 
 mata; in the second part, the development of the function will 
 be considered with reference to the cellular processes by which 
 the glaod takes origin in the guinea-pig, and their parallelism with 
 the cellular processes in the functional periods of the mature 
 organ. 
 
 From the observations of Mr G. Bennett, made on the livinor 
 animals in Australia, there seems little doubt that the gland of 
 the Ornithorhynchus does secrete some kind of milky fluid at a 
 certain time of the year. Mr Bennett has seen the milk exudinsf 
 on the skin, in a manner which reminded him of Owen's description 
 of the exudation of the mercury that he had injected into the 
 substance of the follicles from their free extremities. He records, 
 however, the following exceptional case : " This female had evi- 
 dently just produced her young, and the uterine organs exhibited 
 no appearance of any more being likely to be brought forth. 
 The mammary glands on each side were very large; but it is a 
 curious and rather interesting circumstance in the economy of 
 this animal, that after having been shot, no milk could be 
 expressed from the glands. This was the more surprising to 
 me, as the glands were very vascular on the surface, the mammary 
 artery ramifying over them in a most beautiful and distinct 
 manner. The fur still covered that portion of the integument on 
 which the ducts terminated, and there Avas no appearance of a 
 projecting nipple. In the animals which I have subsequently 
 seen with a lacteal secretion, there has been no projecting nipple, 
 and the fur is not even invariably found quite rubbed off at the 
 situation where the ducts of the gland have their termination\" 
 
 The gland is subject to the same periods of activity and rest 
 as in the higher mammals; but the utilisation of its secretion is 
 
 ^ Transactions of the Zoological Society, Yol. i. p. 251. 
 
Cii. v.] OF THE MAMMARY FUNCTION. Ill 
 
 so obviously uncertain that one is entitled to dismiss altogether 
 the teleological idea of a secretion, and to consider the activity of 
 the organ rather as being correlated to that of the other sexual 
 organs, and ultimately to the season of the year. The full expan- 
 sion of the organ is found only in the specimens procured about 
 the month of December; Mr Bennett's specimens shot in August 
 had the gland in the state of extreme contraction, and it is 
 generally found in that condition in the specimens that are sent to 
 this country. 
 
 The gland in its contracted state has perhaps less than one- 
 tenth the volume of the state of full expansion. The hundred or 
 more independent follicles of which it consists are small oat-shaped 
 bodies of a brown colour, about four or five lines long, and a line in 
 breadth at their middle. They are arranged in the form of a fan, 
 with their necks converging to a narrow linear space under the 
 skin about one-quarter of an inch in length. At that point there is 
 an ill-defined break in the jyanniculus carnosus, and the necks, or 
 rather the upper pointed ends, of the follicles appear to lose them- 
 selves in the subcutaneous tissue. When the fur which covers 
 that region — equally with the surrounding skin — is completely 
 scraped off, the skin to which the follicles converge shows, under 
 a lens, no obvious differences from the rest of the skin. The 
 mouths of ducts cannot be detected, nor any other openings than 
 those of the hair-follicles. In microscopic sections through the 
 whole thickness of skin and through the length of the attached 
 follicles as well, no passages resembling ducts are found traversing 
 the skin to its surface. I have made sections of the involuted 
 gland in a direction exactly perpendicular to the skin, throughout 
 almost the whole region where the follicles were attached. The 
 collected ends of the follicles, owing to their brown colour, could 
 be seen, with the naked eye, to end in the corium like a rounded 
 papilla, the follicles in the centre of the convexity coming nearest 
 to the surface. Under the microscope I could trace no obvious 
 communication with the skin from any one of the numerous 
 follicles that the section encountered. Ducts with an epithelial 
 lining could hardly have escaped notice, although the gland, from 
 long keeping in spirit, did not afford very reliable histological 
 appearances. So far as the follicles of the shrunken or involuted 
 gland are concerned, they appear simply to end in the corium 
 
112 THE DEVELOPMENT [Pt. T. 
 
 with pointed processes. In the absence of better histological 
 evidence, it may be stated as at least probable that the discharge 
 of the secretion on the surface depends not on the presence of 
 permanent ducts, but rather on a kind of transudation through 
 the skin, caused by the accumulation of fluid products in the 
 organ during its period of extreme expansion \ 
 
 The conchision arrived at in the last chapter was that the 
 homologue of the breast is not a cutaneous gland, or, at any rate, 
 that it is not a recess of the skin ; but that it is a fat-body. The 
 primitive condition of the breast tliat is found in the Monotremata 
 may be described, without greatly straining the facts, as a fat- 
 body whose lobules converge to a particular limited area of skin. 
 The follicles of the gland in the female Echidna have, in their 
 state of full expansion, the most curious resemblance to fat-lobules ^ 
 There is the special featui-e of a central passage in each follicle, 
 and round it there are grouped the subdivisions or compartments 
 of the structure. Each of the compartments has its walls as it 
 were honey-combed with saccules or vesicles, which differ from 
 the vesicles of fat-tissue chiefly in their being often much smaller, 
 
 ^ In dissecting a male specimen of Echidna hystrix, which was placed at my 
 disposal by Mr F. M. Balfoiir, I discovered, in a cleft of the panniciilus carnos^is 
 towards the lower end of the abdominal wall on each side, a well-marked organ 
 having the appearance of the involuted mammaiy gland of the female in miniature. 
 The follicles were of the same shape and colour, and converged to a point in the 
 subcutaneous tissue. I made sections, perpendicular to the skin, through the 
 whole group of follicles and attached piece of skin, hut found no trace of open 
 communications between the pointed ends of the follicles and the cutaneous surface. 
 
 On examining the corresponding part of the abdominal wall in the male Orni- 
 thorhijuchus, I found a still more rudimentary form of gland, consisting of only 
 a few of the oat-shaped follicles. The existence of the mamma in the male 
 monotreme has been overlooked hitherto, and Owen has based an argument on its 
 assumed absence. In writing on the subject of this chapter in the Journal of 
 Anatomy and Physiology, Vol. xi. p. 29, I took for granted that the mamma was 
 absent in the male, and endeavoured to show, by a reference to the form of the 
 mamma in male cetaceans, that the glandula femoralis of the male monotreme was 
 the homologue of the mammary gland of the female. A speculation to the effect 
 that the mamma was first "acquired" by the male animal was based upon the 
 supposed homology. The theory and speculation naturally fall to the ground, as 
 theories and speculations often do. I hope to include an account of the mammary 
 glands of the male monotremes in a subsequent paper, along with a number of other 
 new points relating to the glandular system of those animals. 
 
 - I am indebted to Professor Flower for the opportunity of examining this gland 
 microscopically. 
 
Cii. Y.] OF THE JfAMMAIiY FUXCTION. 113 
 
 though many of them are as large as a well-filled fat-cell. The 
 larger vesicles or saccules appear to have membranous walls, or to 
 be divided from one another by membranous septa ; the smaller 
 vesicles suggest rather drops of fluid, and the central space of the 
 compartment is often occupied by a vesiculated fluid. From its 
 staining by logwood, the fluid appears to be of an albuminous 
 or mucous nature. The gland from which the sections were made 
 had been preserved for a long time in spirit, and was not well 
 adapted to show certain histological points, such as the presence 
 of nuclei. 
 
 Those resemblances between the expanded gland of the mono- 
 treme and what may be called the usual condition of fat-tissue 
 are by no means slight. But it is more striking to find that the 
 differences between the gland in its expanded state and in its 
 contracted state are the same as the differences between the 
 fat-body, say of a frog, in its spring condition and in its winter 
 condition. When the contracted or resting state of the gland of 
 the monotreme is compared with its expanded state, the difference 
 is found to be not only in the volume of the organ, but also in 
 the character of its component cells. The compartments into 
 which the lobule or follicle is subdivided are retained in a reduced 
 shape, and the septa between them appear relatively thick. But 
 the cells that lie in the compartments are not the large vesiculated 
 cells, or the honey-comb aggregates of membranous saccules, but 
 round granular and nucleated cells, without a cell-wall, and filled 
 with yellowish-brown pigment. Each space of the follicle con- 
 tains only a few cells, and they lie in the space without any 
 regular arrangement. 
 
 The differences between the cells of the organ in its expanded 
 condition and in its state of contraction resemble the differ- 
 ences betwg^en the ordinary fat-cells and the so-called atrophied 
 fat-cells. The amount of pigment that is dissolved in the fat 
 varies in different animals, and the colour may range from yellow 
 to red, and among the invertebrata to more unusual colours. 
 When the fat disappears from the cell, the pigment remains 
 behind in the cell-substance in the form of a granular mass\ 
 The atrophy and expansion of fat-bodies, or of the fat in certain 
 regions, is a periodical process which generally follows the sexual 
 1 Toiat, Lehrbuch der Gcivebelehre, 1877, p. ii. 
 
 c. c. 8 
 
lU THE DEVELOPMENT [Pt. I. 
 
 periodicity of the animal, tbat is to say, the periodicity of the 
 seasons. In the last chapter I quoted the opinion of Toldt as to 
 the essentially glandular character of fat-bodies; their periodicity 
 is another point that brings them into comparison with the 
 accessory sexual glands. They differ from ordinary secreting 
 glands in so far as they have no mechanism for the discharge of a 
 secretion ; and the mammary glands of the Ornithorhijnchus differ 
 in the same way, though in a less degree. In the latter organ, 
 the system of passages within the gland is of the most primitive 
 kind, the communication between the cavity of the follicle and 
 the surface of the skin is not at all obvious and may be nothing 
 more than occasional, the skin of the mammary region is not even 
 denuded of its fur, and the collection or utilisation of the secretion 
 is manifestly precarious. From all these considerations it appears 
 to me that the mammary glands of the OrnithorhyTichus may be 
 described with equal justice as somewhat specialised fat-bodies. 
 The specialisation must be sought for in several points, but the 
 earliest and least intelligible step in the process of adaptation is 
 the centripetal grouping of the follicles or lobules. 
 
 In considering the development of the glandular function 
 through the line of a homology with fat-bodies, the chief physio- 
 logical factor must be held to be that of periodicity. It is the 
 periodical expansion of the fat-lobule that brings with it the fluid 
 contents of the cells, and it is the j)eriodical expansion of the 
 mamma in the monotremes that issues in the formation and dis- 
 charge of a secretion. Whatever may be the ultimate causes of 
 periodicity within the organism, it appears sufficiently obvious 
 that, both in plants and animals, their periodicity depends in a 
 general way upon the round of the seasons. As periodicity appears 
 to be the most primitive character of certain physiological pro- 
 cesses, in like manner may the presence of this grand factor be 
 traced in the development of function in the individual embryo. 
 The developmental process of the mamma is itself the first 
 functional period of the organ. I shall occupy the rest of the 
 chapter with an exposition of this statement. 
 
 I have shown, in the first and second chapters, that lactation 
 always follows upon a slow unfolding of the secreting structure 
 and a gradual rising of the secretory force. Lactation is the ri.sing 
 secretion maintained for a longer or shorter period at its highest 
 
Ch. v.] OF THE MAMMARY function: 115 
 
 point. If a full period of the breast's activity be represented 
 graphically, a line sloping upwards would represent the evolution, 
 a line continued horizontally would represent the lactation, and a 
 downward extension of the line would stand for the upfolding and 
 subsidence. Lactation is the function of the organ continued for 
 a time from its highest point at a uniform level ; it is a period of 
 variable length interposed — one may say, artificially interposed — 
 between the original waxing and waning periods of the organ. It 
 is not the unfolding and upfolding processes that are subordinate, 
 but it is the period of lactation itself. The original law of 
 periodicity is that "from hour to hour, we ripe and ripe, and then, 
 from hour to hour, we rot and rot." 
 
 Turning now to the formation of the breast in the embryo, as 
 described for the guinea-pig in the fourth chapter, the cells of 
 the embryonic matrix in assuming the secreting structure will 
 be found to exhibit a cycle of changes, which may be compared 
 with the successive changes of a periodical evolution. The steps in 
 the development of the acini are much more rapid than the 
 prolonged stages of unfolding in the mature organ, and the 
 parallelism cannot be observed at every point. But the beginning 
 and end of the process are the same in both cases. The acinus, 
 in the embryonic development, begins as a small group of em- 
 bryonic cells, at first almost nuclear, but afterwards invested with 
 a quantity of cell-substance. The next obvious change is a vacuo- 
 lation of one or more of the group of cells, whereby the circular 
 boundary of the acinus cavity is determined. At a later stage, all 
 or nearly all the cells of the developing acinus are found to be 
 vacuolated to such an extent as to resemble fat-cells. That con- 
 dition corresponds to the highest point in the process of develop- 
 ment of the acini. The vacuolated cells burst their membranous 
 walls and discharge their contents; the discharged contents are 
 the milk of the newborn. All the steps in that process are steps in 
 the formation of the gland ; the issue of the formative process is 
 a discharge of milky fluid. The consummation of the process, 
 indicated by the discharge of a fluid, corresponds, in some animals 
 at least, to the time when the foetus leaves the womb. The dis- 
 charge of a milky fluid from the breasts of newborn children of 
 both sexes is a very familiar observation, and I have observed the 
 same occurrence at or a few days after birth in the animal to 
 
 8—2 
 
IIG THE DEVELOPMENT [Pt. I. 
 
 which my description applies, viz. the guinea-pig. The fluid 
 expressed from the nipple of the new-horn guinea-pig had the 
 appearance of a Avatery kind of milk ; on microscopic examination, 
 the milk-globules of ordinary milk were not found, but a more 
 uniform fluid mass irregularly broken up under the cover-slip into 
 large or small drops, and without any mixture of cellular elements. 
 The ducts of the prepared gland often contained the same kind of 
 fluid. After the acini have discharged the milk of development, 
 they appear to contract their circuit greatly, and in young guinea- 
 pigs at the beginning of their first pregnancy the acini are found 
 in the same resting condition as in the adult, that is to say, they 
 are small round spaces filled with a number of nuclei. 
 
 As the first pregnancy advances, the acini undergo the unfold- 
 ing process that has been described in the second chapter, and 
 when the term of pregnancy is ended, a fluid product is again 
 discharged, this time in large quantity, and continuously for days 
 or weeks as the case may be. It follows from these undoubted 
 facts that the development of the breast is its first period, or that 
 the functional periods of the mature breast are successive repeti- 
 tions of its development. The conclusion is the same that is con-, 
 tained in one of the forgotten aphorisms of Goodsir. "We require 
 renewed observations," he remarks, " on the original development 
 of glands in the embryo. From the information we possess, how- 
 ever, it appears that the process is identical in its nature with the 
 growth of a gland during its state of functional activity\" 
 
 I have already endeavoured to show that the continuous and 
 useful production of milk, which alone constitutes the function 
 from the physiologist's point of view, is not the essential or 
 primary activity of the organ, but a mere prolongation of its 
 evolutional force at its highest point. It is the periodicity of the 
 organ that is its earliest characteristic, and that periodicity, as I 
 make out, is the same as the expansion of a fat-body from season 
 to season. The periodicity of the breast begins with the existence 
 of the breast as a distinct organ ; its development is its first 
 period, and its subsequent periods are all repetitions of its develop- 
 ment. 
 
 One may attempt to analyse, in various directions, the de- 
 velopment of function as distinguished from the development of 
 ' Aiuttoviical (utd FatJiolofiical Ohservationa, Edinburgh, 1845, p. 31. 
 
Cii. v.] OF THE MAMMARY FUNCTION. 117 
 
 structure. But that inherent union of structure and function, 
 which is the grand characteristic of living things, is a dualism that 
 cannot be resolved even in the study of development. In speak- 
 ing of the periodical unfolding of the breast's structure and the 
 correlated revival of its function, I remarked that one might find 
 therein at least an illustration, although not an explanation of the 
 union between structure and function. The one varied as the 
 other varied ; a low degree of function resided in what might be 
 called a crude shape of cell, a high degree of function resided in 
 Avhat might be called a perfect shape of cell, and the intensifying 
 of the function and the perfecting of the cell went band in hand. 
 In like manner the highest point or the last step in the develop- 
 ment of the breast coincides with the discharge of a product, 
 which is the type of its perfect secretion. The first fluid product 
 of the breast is incidental to its development. Questions of 
 heredity apart, the development is an ultimate fact ; while secre- 
 tion is a teleological idea. That which is incidental to the develop- 
 ment becomes the secretion. The milk of development in the 
 breast is the pledge of its future usefulness. 
 
PART 11. 
 
 PATHOLOGICAL. 
 
CHAPTER VI. 
 
 PATHOLOGICAL PROCESSES OF THE BREAST. 
 
 In the first chapters I have described with consideraLle mi- 
 nuteness the various processes or modes of activity that are 
 at work in the mamma of the female throughout the period of 
 sexual maturity. So far as relates to the individual secreting 
 elements of the gland, or to the behaviour of its cells as indi- 
 viduals, the physiology of the breast has been described with a 
 certain amount of completeness. The cellular aspect of the physi- 
 ology of an organ is never without its general interest ; but for the 
 morbid physiological action of the organ, that interest becomes 
 paramount. It is only in a limited department of pathology, in 
 the cise of diseases such as gout and diabetes and the febrile 
 state, that the chemical and physical subdivisions of physiology 
 have a direct bearing. The gi-and disease of the breast is the 
 tumour disease ; and for the rational interpretation of the various 
 departures from the normal of the organ that are included under 
 that name, the account that has been given in previous chapters 
 may be used as a sort of physiological key. 
 
 First among the physiological points that have been discussed I 
 is the fact of the periodical evolution and involution of the gland. ' 
 No other organs of the human body, except the uterus and ovary, 
 are subject to periodical changes of the same kind. The function 
 of the breast intermits regularly for a longer or shorter period ; its 
 subsidence is gradual, and its re-establishment very much slower. 
 In both of the processes the ordinary functional excitation of the 
 gland is still present ; but it acts at a lower degree of intensity, 
 gradually decreasing to complete cessation in the involution, and 
 
122 PATHOLOGICAL PROCESSES [Pt. II. 
 
 gradually rising to its full force in the evolution. In those two 
 inverse processes there is a complete parallelism. They may be 
 both divided into tolerably definite stages which correspond 
 exactly in an inverted order. The stages are characterised by 
 secretory products of the epithelium peculiar to them ; and the 
 peculiar secretory product of each stage, occurring as it does at 
 corresponding periods of both processes, may be considered to be 
 the characteristic product of the excitation of the gland acting at 
 a certain absolute degree of intensity. When the functional 
 stimulus of the mamma is acting at its lowest point, the secretory 
 product is a large granular pigmented cell ; if the mammary exci- 
 tation were always to act at that degree of intensity, the secretion, 
 it may be said, would be always in the form of large granular 
 pigmented cells. At the next appreciable advance in the in- 
 tensity of the stimulus, the product formed in the gland may be 
 described somewhat generally as a large nuclear cell. Beyond 
 the medium degree of intensity, the product is both fluid and 
 cellular, the former being mucus and the latter, generally speak- 
 ing, a small round cell. Coming still nearer to the full excitation, 
 the cellular ingredients are fewer and the mucous production 
 much more abundant. Finally, when the stimulus is at its height, 
 the mucous fluid has given place to a fatty fluid, and whatever 
 cellular elements the secretion contains are the well-known 
 colostrum cells. This summary will be found to be strictly in 
 agreement with the facts already detailed at length. It is sub- 
 mitted again that the parallelism between the involution and 
 evolution processes is not at all strained, and that it is a legitimate 
 inference to ascribe, in the abstract, a special kind of secretory 
 product to a certain degree of intensity of the glandular force. 
 
 The various cellular and fluid products of the imperfect secre- 
 tion are spoken of as waste products. The elaborate mechanism 
 for the disposal and utilisation of the waste cells has been de- 
 scribed in the third chapter. 
 
 Finally, and as a point of more theoretical interest, the law of 
 the secretion, the modus operandi that is at all times implicitly 
 present, and which is clearly manifested in certain of the imperfect 
 states of the function, has been taken to be the law of endogenous 
 cell-formation. It is with reference to the last mentioned and 
 more theoretical of the conclusions drawn from the inquiry into 
 
Ch. VI.] OF THE BREAST. 123 
 
 normal processes, that something remains to be said before passing 
 to the diseased states of the breast. 
 
 The morphological differences between the endogenous mode 
 of cell-formation and the mode by fission or division are on the 
 whole well marked. Cell-formation by fission is essentially a par- 
 titioning of the whole parent cell into two or more ft-actions, which 
 grow and acquire the same form and properties as the cell from 
 which they came. Endogenous cell-formation is the production of 
 one or more young cells within the body of the parent cell ; the 
 young cell rounds itself to a separate whole within a vacuole 
 which contains also a certain amount of fluid substance. Cell- 
 formation is the most elementary phenomenon of living things ; 
 and if there are varieties of cell-formation, it cannot be but that 
 these varieties have a deep significance. Perhaps the two modes 
 of cell-formation are not radically contrasted at all points ; there 
 are certainly phenomena of cellular activity that might be claimed 
 for either of them. But the occurrence of ambiguous cases can- 
 not be held as obliterating the distinction between them ; and, on 
 a broad survey of the facts, it seems admissible to conclude that 
 well-marked varieties of the process exist. 
 
 Such then being the case, it will become, on consideration, a 
 matter of some surprise that more has not been made of the 
 varieties of cell-formation in the details of biological study; for 
 variations in so fundamental a matter as the origin of cells must 
 have a meaning. There is one particular laxity of thought and 
 expression, running through many pathological writings and not 
 altogether absent from biological, which is an example of the way 
 that fundamental varieties of cellular activity have been over- 
 looked. One of the most usual forms assumed by cells subject to 
 the endogenous activity is that of a signet-ring. The cavity of 
 the cell may contain nothing but fluid, the cellular element being 
 on the periphery; and I have lately observed in preparations made 
 from the granulation tissue covering a stump, that the periphery 
 of the cell, corresponding to the stone of the ring, has contained 
 not one nucleus only, but an enormous number of small nuclei. 
 In other cases, the young cell or brood of cells lies within the 
 cavity or vacuole of the parent cell. It so happens that the cells 
 of ordinary fat-tissue exhibit the same appearances, chiefly in their 
 development but to some extent also in their mature condition. 
 
124 PATHOLOGICAL PROCESSES [Pt. II. 
 
 The fat-cell has come to be regarded by many writers as the type 
 of all such appearances, occurring under whatsoever circumstances ; 
 and it has often happened that vacuolated cells, exemplifying the 
 endogenous process, have from their resemblance to the common 
 and, as it were, contemptible fat-cell, been summarily dismissed as 
 examples of "fatty degeneration," It may be safely asserted that 
 a large proportion of cells so described are neither fatty in their 
 chemistry nor degenerating in their vitality\ If the practice were 
 abandoned of referring all such vacuolar transformations of cells 
 to one single and simple category of degeneration, there would be 
 a better prospect of arriving at a true and rational generalisation 
 of these manifold appearances. 
 
 Among pathological writers, Virchow has assigned an import- 
 ance to the varieties of cell-formation that seems more adequate 
 to the case. Under the name of Physaliphoren he describes and 
 figures the usual varieties of vacuolated cells undergoing the 
 endogenous process, and he applies that special name in order to 
 distinguish them from cells, such as amoeboid cells, that exhibit 
 transient vacuolar appearances. The significance that I shall 
 ascribe to the endogenous mode of cell-formation in the patho- 
 logical considerations that follow, is the same that may be found 
 in Virchow's writings, and is suggested by them. Contrasting the 
 hyperplastic and the metaplastic or heterologous modes of growth, 
 he observes, as regards endogenous new formation, that heterology 
 characterises it from the outset, inasmuch as the elements pro- 
 duced in the mother-cell are as a rule small, apparently "in- 
 different " like granulation-cells, and predisposed to a devious 
 course of development^ The property of metaplasia may be con- 
 sidered as belonging to the endogenous product under all circum- 
 stances ; it may be said to inliere in all cells produced according 
 to the law of endogenous cell- formation ; and if we are content, 
 for brevity's sake, to use a formula for the expression of the con- 
 tinuous series of processes in the breast that are not all obviously 
 
 ^ As regards actual fat-cells, or the cells of fat-tissue proper, a correspoudiug 
 fixedness of idea prevails, and sometimes proves curiously inconvenient. Thus 
 Flemming, while he correctly states the endogenous character of the phenomenon, 
 describes the presence of a brood of young cells within the cavity of a fat-cell, and 
 the associated disappearance of the fat, as an example of "growth-atroi)liy" 
 {Wuchcr- Atrophic). Paradox can go no farther. 
 
 * CcUular-l'diluilnnic, p. ■i\}2. 
 
Ch. VI.] OF THE BREAST. 125 
 
 endogenous, we are at once led to the conclusion that all the 
 cellular products of the secretion, or, in other words, the waste 
 cells of the secretion possess the same property of metaplasia. It 
 is in this property of metaplasia that their importance as regards 
 disease chiefly consists. For metaplasia is nothing else than the 
 heteroplasia which we uniformly associate with the " parasitism " 
 or malignancy of tumours. 
 
 The tumour diseases described in this and the following chapters 
 have been studied chiefly in the bitch. During a period of three 
 years, I collected material from upwards of twenty cases of mam- 
 mary tumours in the bitch, that were under treatment at the 
 Brown Institution (hospital for animals), London \ I obtained 
 also two mammary tumours from the cat. Although the tumours 
 in the dog and cat differed in some respects from the ordinary 
 mammary tumours in the human body, they had, as a class, the 
 property of malignity equally well marked. Thus, in one of the 
 cases in the cat, the tumour occurred as a circumscribed mass at a 
 particular part of the chain of glands, and was removed by opera- 
 tion ; there appeared, soon after, a fungus-like growth in the axilla 
 and the animal died with numerous secondary tumours in the 
 lungs. Several of the cases in the bitch ended fatally, and proved 
 to have infection of the lymphatic glands with tumours in the 
 lungs or liver. Most of the cases were operated on, and were not 
 afterwards heard of; but three of the animals came under treat- 
 ment a second time, the disease having broken out afresh, and 
 over a larger extent of the gland. In about one-third of the cases 
 in the bitch, there was the well-known tendency to the formation 
 of cartilage at certain points ; and in one case a large part of the 
 principal tumour, but no part of the smaller nodules, was distinctly 
 osteoid. But although a certain proportion of the cases showed a 
 tendency towai'ds the comparatively safe cartihiginous transforma- 
 tion, that tendency did not always exempt them from malignity. 
 Thus, in an extensive tumour, removed from an animal that had 
 undergone the same operation eighteen months before, there were 
 found several small cartilaginous points ; elsewhere the structure 
 was of somewhat soft consistence, and the lungs were full of 
 
 ^ I am greatly indebted to Mr William Duguid, the veterinary siirgeon of the 
 Institution, for giving me the opportunities, from time to time, of making the 
 collection of tumours, and for assistance in preserving them. 
 
126 PATHOLOGICAL PROCESSES [Pt. II. 
 
 secondary tumours having the structure of the softer kind of tissue 
 in the primary. A certain myxomatous character, of which the 
 cartilaginous tendency may be considered to be a further develop- 
 ment, was very generally present; one tumour (in a case where there 
 were secondary growths in the liver) had a large cyst on its upper 
 surface, filled with colloid fluid, and several of the cases differed 
 in appearance from skirrhous cancers of the human subject chiefly 
 in the brownish colour of the tumour juice. If the tumours of 
 this collection were referred to their respective classes, there would 
 be representatives among them of skirrhous cancer, medullary 
 cancer, reticular cancer, alveolar cancer, cystic sarcoma, &c. I 
 have introduced a brief account of eight cases of mammary 
 tumours in the human subject, chiefly with the object of showing 
 that they illustrate the same principles of tumour formation that 
 I endeavour to establish from a more particular study of the cases 
 in the bitch. 
 
 The cases in the bitch have the great and unique advantage 
 for study, in that the tumour disease is often found in various 
 stages at various points of the elongated chain of glands. Thus, 
 in a particular case which ended fatally, there was one large 
 tumour about two inches in diameter, and numerous small circum- 
 scribed nodules throughout the wliole length of gland on each 
 side ; and, at one part, there was not so much a circumscribed 
 tumour, but rather a uniform enlargement or thickening over the 
 whole breadth of the gland and for a distance of about two inches 
 longitudinally. In the animals that died or were killed I was able 
 to secure the whole chain of glands with their numerous centres 
 of morbid action; and in the tumours removed by operation I 
 found, in almo.st every case, considerable fringes of more or less 
 normal mammary tissue, which, though more troublesome to pre- 
 pare, were hardly less valuable for the pathological inquiry than 
 the entire chain of glands would have been. The material of the 
 investigation included, therefore, not only tumours of the mamma, 
 but also portions of the mammary tissue illustrating every stage of 
 the tumour disease. 
 
 The attemjDt to refer the mammary tumours of the bitch to 
 their respective classes, is here deliberately abandoned as serving 
 no good purpose. It will doubtless be found to be more in keeping 
 with the etiological end proposed for the investigation, if the 
 
Ch. VI.] OF THE BREAST. 127 
 
 attempt be made to find out what are the various starting-points 
 of the diseased process from the normal. The first part of the 
 inv'estigation supplies, as has been said, a physiological scheme 
 whereby to trace out and to estimate the extent of these abnor- 
 malities. The fringes of more or less normal gland substance, and 
 the younger tumour nodules or enlargements form a kind of inter- 
 mediate territory, in which the traces of healthy action are not 
 altogether hopelessly obscured. It is in dealing only Avith this 
 kind of material that one can hope to indicate the exact point of 
 departure of the disease. There is indeed an infinite variety of 
 appearances in the pathological mammary structure ; there is 
 nothing of it, one may say, but is changed " into something rich 
 and strange." But those vagaries of nature may be reduced to a 
 few leading principles. In the first part of the investigation we 
 profess to have established one or more important laws of the 
 healthy mammary activity, and the whole of the pathological 
 investijration that follows will be found to be nothinsr else than an 
 illustration of the working of those laws under altered circum- 
 stances. It is proposed, therefore, to give in the first instance a 
 number of illustrations of morbid physiological action within the 
 breast, and to point out from time to time the relation of the same 
 to the various kinds of tumours. 
 
 The most common appearance in the more or less healthy 
 mamma adjoining tumours, and in certain parts of the circum- 
 scribed tumours, is an appearance resembling that of the eaidy 
 evolution of the mamma. The figure No. 21 is taken from the 
 mamma near a tumour. It is not at all different from the breast 
 in the early stage of evolution, but there was no reason to suppose 
 in this or in any other of the cases that the animal was subject to 
 the ordinary cause of the unfolding of the breast from the resting 
 state, viz. pregnancy. The effect, however, is the same ; the acini 
 are obviously subject to a certain degree of activity; they contain 
 a number of the large yellow cells in their cavity and there is a 
 considerable number of the same large yellow cells in passage 
 through the loose fibrillar tissue outside. This appearance is con- 
 stantly found in the bitch as the commencement of disease, and 
 the process so far is a diseased process because the excitation to it 
 is something other than the natural excitation to the evolution of 
 the mamma. The gland is disturbed from its resting state and 
 
128 
 
 FA THOLOGICA L PROCESSES 
 
 [Pt. II. 
 
 proceeds to follow, to a certain point at least, the ordinary course 
 of evolution. But, if one may so speak, it has been started along 
 its wonted evolutional track on a false pretence. The evolution is 
 spurious, and the end of it, as will be seen, is disastrous. Strictly 
 
 Fio. 21. 
 
 Lobule of a ruamma near the resting; state. Numerous lai-pe pigmented cells within 
 the acini and in the interlobular fibrillar tissue; sections of three blood-vessels to the 
 left of the lobule. Masnifled 160 diameters. 
 
 speaking,, however, the process is not one of evolution ; it is rather 
 the absolute effect of a certain low degree of excitation acting 
 npon a gland in the resting state. 
 
 As we have already seen, the characteristic product of the 
 lowest degree of excitation, whether at the end of involution or at 
 the beginning of evolution, is the large granular pigmented cell. 
 In the involution process, the excitation speedily comes to an 
 end ; the production of the large yellow cells is a passing pheno- 
 menon which the subsidence of the function gives few oppor- 
 tunities of observing welL In the evolution process it can be 
 studied very much better ; the cells are often found in situ within 
 the acini, and large accumulations of them may be seen outside. 
 But in the evolution also the phenomenon is transitory; the 
 degree of the excitation is gradually increasing, and, for that or 
 other reasons, the large yellow cells cannot be easily seen in what 
 we must consider to be the successive steps of their career. In 
 the cases of spurious excitation, however, the circumstances admit 
 of the mode of production of the yellow cells being much more 
 
Ch. VL] 
 
 OF THE BREAST. 
 
 129 
 
 deliberately observed. In the diseased process we realise partially 
 that state of the secretion, corresponding to the third of Goodsir's 
 " three orders of secretion " in the lower animals, where it does not 
 advance to the production of milk, but continues all the time to 
 act at a uniform low degree. The characteristic yellow cell is 
 liere a much more stable product, and, as will appear, sometimes 
 
 Fkj. 22. 
 
 1 «»!>-, 
 
 ffx 
 
 
 ^^ 
 
 
 
 A pathological lobule. The uppermost acinus shows the 
 p'pmeiited cells as epithelium in situ. Magnified 160 
 diameters. 
 
 accumulates to form an epithelium of several layers. In certain 
 cases the whole floor of an acinus has a yellowish tinge, a certain 
 amount of pigmented substance being disposed round each nuclear 
 cell. The acinus seen also in profile is found to have a complete 
 circlet of yellow cells. Fig. 22 is a drawing of a portion of a 
 lobulus of the mamma near a tumour. The lobuli near it are at 
 c.c. 9 
 
130 
 
 PA TlIOLOaiCA L PROCESSES 
 
 [Ft. II. 
 
 or about tlie usual resting state ; they arc smaller than tlie one 
 figured, and separated from one another by the usual amount of 
 fibrillar tissue. The whole of this pathological lobule has a dis- 
 tinct yellow tint, and the yellow colour of the cells in the half 
 acinus at the top of the figure is very well marked. Their sub- 
 stance is the ordinary granular pigmented substance, but their 
 form is peculiar. It is neither the round form of the normal pig- 
 mented cell, nor the polyhedric form of the perfect epithelium ; 
 but it is more of a columnar or club shape. Most of the cells in 
 the other acini of the lobule belong to a class that will fall to 
 be described afterwards. Fiff. 23 shows a form of cell somewhat 
 
 Fig. 23. 
 
 pf 6 J^\ 
 
 ^ / 
 
 From the mamma of a bitch, adjoining a tumour. Tho pjiithelium massed in 
 more- than one la.ver, or in villus-like processes. The granular substance of the 
 cells is everyivhere brir/ht yellotv. Maguifled 300 diameters. 
 
 similar to that in the upper acinus of Fig. 22. The preparation 
 is taken from the very much affected gland tissue round a tumour, 
 and the striking point of the appearance is the brilliant yellow 
 of the entire semicircle of cells. Where the cells are lying single 
 they are found to be more or less columnar ; elsewhere they are 
 fused together into villus-like processes, consisting of a double 
 row of three or four columnar cells on end. A more remarkable 
 degree of the same kind of aggregation in situ of the pigmented 
 cells is seen in Fig. 24. This is taken from the tolerably healthy 
 gland tissue near a tumour. The portion represented in the 
 
Ch. VI.] 
 
 OF TEE BREAST. 
 
 131 
 
 figure is evidently the recess of a duct, tlie epithelium of which 
 shows the same characters as may be seen elsewhere within the 
 acini of the gland. All the substance that is made to appear 
 granular in the woodcut is in the preparation of a bright yellow. 
 
 Fig. 24. 
 
 
 
 ^v^j«®^|^ 5ji^' , 
 
 w./ 
 
 ■-i^':':^ 
 
 From the mamma of a bitch, adjoining a tumour. Numerous viHus-like proc sses of 
 epithelial cells, ^owins; from the recesses of a duct. The cells are all pigmented yellow. 
 Magnified 150 diameters. 
 
 The two perpendicular sides of the recess have a border of yellow 
 granular substance, which can be seen clearly enough at most 
 places to consist of large cubical cells. From the bottom of the 
 recess there springs a cluster of papillary processes, some of which 
 
 9—2 
 
132 PATHOLOGICAL PROCESSES [Ft. II. 
 
 are cut obliquely and appear to be short, while two or three 
 of them are long columns of uniform thickness, built up solely of 
 laro-e yellow granular cells arranged round a narrow central space. 
 In the class of appearances from which Figs. 23 and 24 are 
 taken, there is evidently a great hyperplasia of the epithelium : 
 the growth is almost vegetative in its characters. But the whole 
 essence of the pathological process lies in the fact that there is 
 more than a mere hyperplasia or excessive reproduction of epi- 
 thelium. The yellow granular appearance of the cells is certain 
 evidence that they are subject at the same time to the functional 
 activity of the gland. It has been argued at some length in 
 the physiological chapters that the formative activity of the 
 mamma and the functional activity always go hand in hand. 
 The formation of the secretion from successive sets of the epi- 
 thelium presupposes constant relays of new cells, which, in the 
 greatest activity of the function, are destroyed almost as fast 
 as they are formed. What we have here is the formative ex- 
 citation outrunning the functional. For whatever reason, the 
 epithelial cells are not shed as soon as they undergo their degree 
 of functional transformation, but they remain in situ. They 
 do not form a many-layered epithelium after the ordinary pattern; 
 they have rather a tendency to run out into a number of papillary 
 processes. The reason of their accumulation in situ is probably, 
 as has been already suggested, the fact that the excitation is 
 not the progressive excitation of evolution, but a certain low 
 degree of the functional activity called erroneously into play 
 and actinsf at the same low rate for an indefinite time. The 
 secretory product under these circumstances would appear to 
 be a more stable product than are the A\"aste cells produced by 
 a corresponding functional stimulus in the healthy evolution, 
 and no doubt the increment of the secretory force in the latter 
 case accounts for the difference. The cell is, however, still a 
 product of the secretion ; its yellow colouring is conclusive on 
 that point. It belongs to exactly the same category as the 
 waste cells of the healthy process ; and, to adopt the convenient 
 formula that has been used already, those yellow cells are 
 produced according to the law of endogenous cell-formation. 
 Such products, again, have been explained to be " metaplastic " 
 or heteroplastic. 
 
Cu. VI.] OF THE BREAST. 133 
 
 The large yellow cells are uot the most abundant constituents 
 of the new growths in the bitch — in tliose of the human subject 
 they do not appear to occur often — nor can any considerable part 
 of the morbid structure be referred to transformations of them 
 except in one or two cases ; but they are found in almost all 
 the tumours as an important concomitant of the other processes, 
 and they have a special importance for investigation from their 
 striking colour. They are in fact, as we found also in the physio- 
 logical part, to be taken as representing the waste products in 
 general ; and the explanation of pathological processes in the 
 breast becomes much easier and more objective if they are used, 
 with due caution, in that capacity. Up to this point we have 
 found that, in the spurious excitation of disease, the pigmented 
 cells are produced in excess within the acini. But in most cases 
 it is not inside the acini, but in the fibrillar tissue outside that 
 they are found in greatest numbers. In Fig. 14, illustrating 
 the normal evolution, there is seen a large collection of precisely 
 the same cells in the tissue surrounding the secreting structure, 
 and it has been explained that the large accumulation at one 
 point was in that case the result of an accident. Now in the 
 mammae subject to a diseased process, such collections are con- 
 stantly found. The cells are frequently deposited in several 
 rows between the acini — which are themselves the seat of morbid 
 changes to be described afterwards — but the largest collections 
 of them are in the wide tracts of fibrillar tissue which separate 
 the lobuli in the resting state, and which appear to become 
 much more extensive in the mammae of aged animals, just as 
 the fibrous stroma of the breast becomes the predominant tissue 
 in women at or beyond the climacteric period. In those tracts 
 of fibrillar tissue the pigmented waste-cells are sometimes found 
 in large irregular heaps ; more frequently they are seen to be 
 arranged in certain regular lines, which obviously correspond to 
 the pre-existing lymphatic spaces. In such situations they usually 
 lose their roundness ; they become oblong, with their opposed 
 ends flattened, and not unfrequently they become long, narrow, 
 and fusiform. The nucleus, which is generally visible, takes on 
 a corresponding alteration of shape. Fig. 25 is a high-power 
 drawing of one small portion of a field that is quite covered with 
 the large round yellow cells. It is from the diseased mamma 
 
134 
 
 PA TIIOLOGICA L PROCESSES 
 
 [Pt. II. 
 
 of a bitch, the same from which Fig. 22 is taken. The latter, 
 from the less changed part of the gland, will .serve to show 
 pictorially where the yellow cells in the fibrillar tissue have 
 come from ; the pigmented cells of Fig. 25, which are disposed 
 in groups in the fibrillar tissue and which contribute to the bulk 
 of the tumour, are the same yellow cells that were before seen 
 in situ in the acini of the gland. If the extra-acinous collection 
 be examined closely, it will be seen to be made up of rows of 
 
 Fig. 25. 
 
 
 
 
 
 
 ■i'// 
 
 
 
 
 
 '^^^ffiilL.. 
 
 From a mammary tumour in the bitch. The fibrillar stroma of 
 the mamma invaded by rows and alveolar groups of largo ycUuw 
 granular cells. Magnified 300 diameters. 
 
 cells which unite to form alveolar groups. Often a single file 
 of cells extends for a long distance across the field ; or there may 
 be a double row, or any variety of grouping up to the alveolar 
 collection, half a dozen cells or more in breadth at its broadest 
 part. In the immediate neighbourhood of acini, or in the spaces 
 between them, the arrangement of the yellow cells nearly always 
 follows the circular outline of the acinus. It does not appear 
 that the large granular pigmented cells ever undergo division of 
 
Ch. VI.] OF THE BREAST. 135 
 
 tlieir substance. If the collections of them increase in extent, it 
 is by the addition of ficsh supplies from the original source. But 
 they are found to undergo certain transformations, and the 
 derivatives of the pigmented cell, the cells resulting from such 
 transformation, appear to multiply in the usual manner. The 
 transformation consists most commonly in the loss of their pig- 
 mented cell substance. 
 
 In order to take advantage of the illustrations of extra-aciuous 
 
 grouping of cells, which are atforded by the waste cells of the j^ig- 
 
 mented stage, it is desirable to refer here to a very common type 
 
 of tumour of the human female breast, although its component 
 
 cells may seem to have an entirely different origin. Fig. 26 is a 
 
 Pjq 26. drawing of one of the most common appear- 
 
 Q ances in the intensely hard skirrhus of the 
 
 ;' teP K^'-^f^J ; U il breast of women at or beyond the climac- 
 
 '^ /•^§.'i-0(ji^fi'''^^''^ ^®^"^^ period. In this particular case the 
 
 '^I'^^^iM^s^ui^:^}^ patient was fifty-eight years old. If the 
 
 ■ -'^ r'fj '(-:;_v/;| figure be compared with Fig. 25, the cells 
 
 ' |\ift"' ''; ,:aU''| will be found to be very different; they are 
 
 mi'^\ -- •>,'7j rather less than half the size, they are little 
 
 ^'■^ '*^ ■ " '^Xrh else than nuclei without cell substance, and 
 
 m 
 
 ^^/^l^^\v: te I 1 they have no pigmentation. Those diflfer- 
 %\ 05^ ' :^^ 0y ..' ences are, however, unessential to the ar- 
 From a skirrhous tumour Aliment: they are merely the differences 
 
 of the human female breast. ° -^ , _ "^ _ 
 
 The fibrillar stroma of the botweeU OUO kind of epithelial waste pro- 
 
 breast occupied by rows and '■ ^ 
 
 alveolar groups of nuclear duct and another. The essential point of 
 
 cells. Magnitied 300 dia- ... 
 
 °^*^''^''^- similarity is the presence of cells in the 
 
 interfascicular spaces, and the grouping of the same. The position 
 of the cells in the two cases is precisely alike. A certain number 
 of them are arranged in single file, and these have frequently a 
 cubical or discoid shape from the flattening of their opposed sides ; 
 the greater number of them are grouped in alveoli of varying 
 breadth, and the alveoli are obviously made up by the fusion of 
 two or more rows of cells. In short, if the preparations from the 
 case of skirrhus of the human breast are compared with the pre- 
 parations from many cases of tumours in the bitch, which have the 
 pigmented cells as part of their constituents, the resemblance in 
 the grouj^ing of the cells is found to correspond throughout ; and 
 there arises the strongest presumption that the cells of the skirrhus 
 
136 PATUOLOGICAL PROCESSES [Pt. II. 
 
 are of the same origin and subject to the same influences as the 
 pigmented cells, which are known beyond all doubt to be the 
 waste products of a feeble degree of the secretory activity. The 
 strong presumption that arises at this stage of the inquiry will be 
 supported by a subsequent description of the same nuclear cells as 
 actual derivatives of the epithelium, or as waste products of the se- 
 cretion. In the meantime it is necessary to observe that the cells 
 of skirrhus, although they are in great part situated outside the 
 secreting structures of the gland, or the remnants of the same, are 
 not all so situated ; and further, that it is'precisely in those breasts 
 that have an enormous development of the interlobular fibrillar 
 tissue, that is to say, in the breasts of women at or beyond ths 
 climacteric period, that the skirrhus form of tumour usually 
 occurs. 
 
 The appearance drawn in Fig. 26 is selected, because it is 
 almost exactly the same as that figured by Virchow to prove the 
 origin of the cells in cancer of the breast from the pre-existing 
 connective-tissue cells of the part\ The direct evidence in favour 
 of Virchovv's theory is necessarily inconclusive ; the appearances 
 are ambiguous, and the theory is hardly capable of proof by 
 direct and objective demonstration. But Virchow's theory of 
 the origin of the cells of cancer is only a corollary of one of the 
 doctrines of the cellular pathology. Instead of the universal ex- 
 udation or plastic lymph of the earlier school of pathologists, he 
 finds that the all-pervading connective tissue, the cellular elements 
 of which he himself demonstrated, is the matrix-tissue of new 
 growths*. Virchow's substitution of the cellular elements of the 
 connective tissue for the formative blastema or plastic lymph of 
 his predecessors has long been an accepted doctrine, and the 
 historical value of that great step in pathological knowledge will 
 never be forgotten. But it has become more and more difticult, in 
 view of the numerous contributions to general pathology made in 
 the interval, to maintain Virchow's connective-tissue doctrine in 
 its entirety. Even in so general disorders as secondary infections, 
 whether acute, or chronic, or tumour-infections, it appears to be not 
 so much the all-pervading connective tissue, as the predominant 
 
 1 Cellular-Pathologie, p. 539, Fig. 1^6. See also the diagrams in Virchow's 
 Lectures on Tumours, Lecture V. 
 ■•^ Cellular-Patholof/ie, p. 485. 
 
^^. VI.] OF THE BREAST. 137 
 
 cells of eacli particular region that take ou the infected action, 
 producing secondary abscess, tubercle, or tumour, as the case may 
 be. Still less is it tenable that the connective tissue, all-pervading 
 though it be, may form the matrix of all the primary new groAvths 
 that have been referred to it. In so far as the tissues of the 
 connective series are the matricular tissues of new growths, it is 
 not from any general liability of that kind of tissue, but from a 
 particular liability of the tissue in each particular region. Thus, 
 if the stroma of the breast be really the source of the cells that 
 go to make a skirrhous tumour, that is a local and special activity 
 of the stroma of a particular gland. The stroma of the breast, ac- 
 cording to my conclusion in the fourth chapter, is the same in its his- 
 togenesis as the glandular parts ; and a systematic investigation of 
 the climacteric changes in the breast may be expected to show 
 something like an obliteration of the glandular parts as distin- 
 guished from the stroma, or at least a condition in which the 
 stroma greatly predominated. It is in that condition of the breast 
 that the skirrhous form of tumour is apt to occur. For that par- 
 ticular form of mammary tumour, therefore, the question of its 
 histogenesis is not absolutely as between connective tissue stroma 
 on the one hand and epithelial glandular structure on the other. 
 Skirrhous cancer of the breast is a disease incidental to the eflface- 
 ment of the glandular structure, and to the absolute and final 
 withdrawal of the secretory force. The real point at issue as 
 regards Fig. 26 is whether the cells have developed where they 
 lie, or whether they may be regarded as mobilised cells of the 
 organ or of the remains of the organ ; and the similarity of group- 
 ing in Fig. 25 affords a certain amount of evidence on the one side. 
 The theory of the epithelial origin of such tumours as skirrhus 
 of the breast has often been objected to\ inasmuch as that theory 
 seems to afford no explanation of the heterology of the new growth. 
 Such tumours have no doubt been described as formations of " aty- 
 pical" epithelium^ but it has not been explained wherein the 
 atypical element lies. Besides the word " atypical," there is no- 
 thing in the statements of Waldeyer and other supporters of the 
 theory, to show wherein the new growth differs from a mere hyper- 
 plasia ; their language is merely the language of structural exten- 
 
 1 Virchow, loc. cit. p. 569. 
 
 - Waldeyer, Virchow's Archiv, Vol. 55. 
 
138 PATHOLOGICAL PROCESSES [Pt. U- 
 
 siou ; the new growth depends upon " proliferation," " budding," 
 &c. Oil the other hand, the atypical element or the element of 
 heterology, in Virchow's theory, is explained by deriving the 
 quasi-epithelium of the new growth from the connective-tissue 
 cells. But the heterology of the tumour formation in a secreting 
 organ is sufficiently explained according to an argument that I 
 have already stated. The pathological activity of the breast is 
 not merely in the way of structural growth ; the functional factor 
 is inseparably associated with the structural. The cellular pro- 
 ducts of the gland, whether under normal or morbid circum- 
 stances, are in the nature of things metaplastic ; and if the cellular 
 products of the gland, or the waste colls of the secretion, aggregate 
 to form a tumour, that tumour is in the nature of things hetero- 
 logous. 
 
 Returning from this digression to the illustration of morbid 
 physiological action in the mamma of the bitch, and taking 
 the phenomena in their physiological order, we come next to a 
 certain class of pathological processes that play a far greater 
 part in the actual formation of new growths than the simple 
 but striking waste production of the pigmented sort which was 
 last described. These also have their type in a certain phase 
 of the normal activity of the breast. 'i'hey correspond to a 
 more advanced stage of evolution, or, absolutely, to a higher 
 degree of functional excitation. It is necessary, however, to 
 refer at this point to an elementary fact or law of tumour-for- 
 mation in the breast, which underlies all the varieties of the 
 process. 
 
 In nearly all the cases of mammary tumour-disease in the 
 bitch, the fringes of gland tissue round the tumour, or the glands 
 generally, where the whole of them could be examined, were 
 found to be either in the extreme involuted condition or to 
 show only the first removes from the resting state. The most 
 probable inference from that fact is that the gland became 
 subject to the morbid influence while its structure was upfolded 
 and its function quiescent. The particular portions of the chain 
 of glands where circumscribed tumours or more diffused enlarge- 
 ments are found, are those parts of the organ in which the 
 morbid excitation has begun earlier, or has advanced more 
 rapidly, or has in general met with more favourable conditions. 
 
Ch. VI.] OF THE BREAST. 139 
 
 The various forms of tumours, as the sequel will endeavour to 
 show, correspond to the various states of the secreting structure, 
 and to the various degrees of the functional force, as measured 
 on the physiological scale. Collections of the large pigmented 
 cells, the earliest or lowest gland products, are generally found 
 as concomitants of the tumour-formation, and in some cases 
 those cells or their derivatives are a not unimportant part of 
 the new growth ; cells corresponding to a stronger degree of 
 the functional force are the elements of tumour-formation in an 
 important class of cases to be described immediately ; and in 
 the next chapter, varieties of mammary tumours in the bitch 
 will be discussed, which appear beyond all doubt to correspond 
 to that condition of the secreting structure and to that degree 
 of the secretory force which, in the physiological chapters, was 
 •found to be associated with the production of mucus. It does 
 not appear, however, that the degree of spurious excitation 
 whereby these characteristic varieties of structure are produced, 
 can be attained to by the gland without the earlier and lower 
 degree of excitation, the first departure from the resting state, 
 having preceded it. The breast is an organ that is made subject 
 to periodicity ; taking it at the resting state, it cannot under 
 any circumstances reach the perfection of its function without 
 going through the somewhat slow series of unfolding changes. 
 When the evolution that is set up is of a spurious kind, or in 
 other words, when a gland is disturbed from its resting state by 
 some cause other than pregnancy, the steps of its unfolding are 
 less orderly than in the normal evolution ; and the fatality of 
 the morbid process consists in this, that the spurious excitation 
 never carries the gland to the end of its unfolding, or to the 
 perfect degree of its function. The products of the gland never 
 get beyond the crude condition, and it is the crude or cellular 
 kind of secretory product that makes the tumour. 
 
 But although the facts themselves, as derived from the series 
 of cases in the bitch, go to prove that the diseased excitation 
 attacks the gland in the resting state, and leads it so far through 
 a course of spurious evolution, it is not to be forgotten that there 
 may be interferences with the activity of the gland at other times 
 or under other circumstances. Thus, the characteristic disorder 
 of the breast during lactation, or on a sudden stoppage of lactation. 
 
140 PATHOLOGICAL PROCESSES [Pt. 11. 
 
 is inflammation and abscess. Again, although it may not be 
 possible to obtain precise clinical facts in evidence, it is reasonable 
 to suppose that the upfolding process of the breast at the end 
 of suckling may sometimes take a morbid direction, or the sub- 
 siding force continue as an enfeebled excitation for a length of 
 time\ As regards the prolonged unfolding process during preg- 
 nancy, it is not difficult to imagine circumstances, such as abortion, 
 which may cause the gland, or some part of it, to leave the 
 
 Fig. 27. 
 
 From the mamma of a cat preg- 
 nant with fiKtuses 8i inches loiisr. 
 Nuclear condition of the epitlielial 
 cells. MaKiUlied 300 diameters. 
 
 beaten track of health. Tumour-formation may very well be 
 incidental to the periodical upfoldings and unfoldings of the 
 breast, but there is a want of positive evidence that it is so. 
 
 Premising, then, that most cases of tumour-formation in the 
 bitch begin with those early evolutional changes that have been 
 already described and figured (Figs. 21 to 25), we come at length 
 to treat of actual new growths that have their structural type in 
 a more unfolded condition of acini and in a higher degree of the 
 
 ^ la the case of a woman with cancer of the breast, who bore children up to the 
 age of about forty-tive, I elicited from her a fact that she had specially noticed, 
 viz. that a fluid escaped from the breast for several months after the weaning of 
 her last child. 
 
Cii. VI.] OF THE BREAST. 141 
 
 secretory force. The type of their cells may be said briefly to 
 be that condition of the normal glandular epithelium ■which is 
 represented in Fig. 27, repeated from the second physiological 
 chapter. 
 
 The figure gives the appearance of the mammary acini in 
 the cat at the period of unfolding that follows the production 
 of pigmented waste products, and that precedes the formation 
 of mucus. The epithelium of the acinus at that intermediate 
 stage is in the form of somewhat large and granular nuclear 
 cells. Compared with the perfect epithelial cell, they are all 
 nucleus and without cell substance ; compared with the nucleus 
 alone of the perfect epithelial cell, they are larger and more 
 granular. Without entering again into the question whether 
 they themselves are thrown off from the acini as waste cells, 
 they may at least be taken to be the crude kind of epithelium 
 in which an imperfect secretory force resides. 
 
 Now, although the fact has not been sufficiently noticed by 
 pathological writers and has certainly never been placed in its 
 exact physiological category, the most ordinary cell in mammary 
 tumours of the human subject is this same large nuclear cell. It 
 has been the general practice hitherto to consider the healthy 
 breast alvvtiys as the fully expanded breast, and to take account of 
 no other epithelium of the breast than the perfect polyhedric 
 nucleated epithelial cells ; and pathological writers who have 
 described tumours of the breast as excessive and irregular growths 
 of epithelium have either tacitly ignored the difference between 
 the pathological cells and the polyhedric epithelium, or, having 
 admitted a difference, they have treated the tumour cell (cancer 
 cell) as one of those entities that exist for pathology alone, and in 
 antagonism to health. But the somewhat large and granular 
 nuclear cell of mammary tumours is nothing else than the crude 
 epithelium of the middle period of evolution. There can be no 
 doubt that the large yellow cells described and figured at the 
 beginning of the chapter are the same yellow cells that occur at 
 the earliest stage of the normal evolution of the breast ; and there 
 is just as little doubt as to the identity of the nuclear cells in the 
 respective cases. 
 
 Fiof. 28 is a drawing: of one of several contiguous acini from a 
 diseased mamma (mammary tumour) of the bitch ; with reference 
 
142 
 
 PATHOLOGICAL PROCESSES 
 
 [Pt. II. 
 
 to the concluding statement of tlie last paragraph, the villous 
 processes of cells in the figure may be at once compared with the 
 villous outgrowths of pigmented cells in Figs. 23 and 24, and the 
 exact similarity of grouping in two very different forms of cell 
 cannot escape notice. In the adjoining figure, one side of the acinus, 
 seen in profile, has the nuclear epithelium massed either in more 
 layers than one, or in a number of villous or knob-like processes. 
 The appearance is perfectly characteristic of the whole of that 
 
 Fig. 28. 
 
 
 Prom a mammary tumour in the bitch. The epithelium of the 
 ncitius massed in several layers or in processes. Nuclear condition of 
 the cells. Masnified 300 diameters. 
 
 particular case ; and the process, with its modifications and develop- 
 ments, is one of the most common of the pathological processes with- 
 in the breast that, in the gross, make up a cancerous tumour. Several 
 other cases of tumour in the bitch, both the cases in the cat, and 
 certain cases in the human subject, are directly referable to the 
 same initial process. The acini are for the most part greatly 
 enlarged, so much so in certain cases that the section of the tumour 
 has to the naked eye a reticular appearance depending on their 
 
Cir. Vr.] OF THE BREAST. 143 
 
 prominent outlines. In tliose of them where the process is not 
 complicated, the wall of the acinus, or the floor of it, is seen to bo 
 covered with a number of wart-like prominences ; as many as ten 
 or twelve of these may be seen arising from the floor of a single 
 acinus. The prominences are for the most part simple aggregates 
 of nuclear cells having a distinct nucleolus, and they are built up 
 with some regularity so as to leave a small central lumen. The 
 earlier stage of these prominences is seen to be a narrow solid 
 column of cells; and such piles of cells are simply a modification of 
 the many-layered condition of the lining cells of the acinus'. 
 
 Fig. 29. 
 
 
 From a tumour of tlio liununi female breast. Portion of the wnll of 
 an enlarged acinus, showing Ihe epithelium massed in several layers. 
 Nuclear 2couditiou of the cells. Magnified 300 diameters. 
 
 Fig. 29 is a drawing of a part of one side of an acinus from a 
 mammary tumour in a woman. The entire tumour, wduch was 
 situated at one side of the nipple, of small extent, well circum- 
 scribed and of somewhat soft consistence, was made up of nothing 
 else but this condition of the acini and of the further develop- 
 ments and direct modifications of the same. The tumour is in 
 fact representative of the intra-acinous class of new growths, which 
 correspond most closely to the clinical division of medullary cancers 
 
 ' As an opiDortuuity will uot be found to retm-n to the point, one of the common 
 developments of the intra-acinous papillary processes may be referred to in passing. 
 The lumen that tends to form in the centre of the column of cells is often occupied 
 by blood-vessels. The papillary process then acquires a sort of bi-pennate appear- 
 ance, the hne of the blood-vessels forming the raphe, and the epithelium being 
 arranged regularly along each side of it. The vascularised processes often ana- 
 stomose, in the way that will be afterwards described for trabecular structiu'es 
 within the acini, and there results a network of colimins having an epithelial 
 covering on each side, the original outlines of the acini being in the end obsciu-ed. 
 "When the vascularised pai^iUaiy processes do not anastomose, they project into the 
 acini as if they were imperfect septa. Langhans {Virchow's Archiv, Vol. 58) has 
 described the same appearance, but he attributes it to an actual breaking down of 
 the septum between two acini, as in emphysema of the lung 
 
144 PATHOLOGICAL PROCESSES [Pt. II. 
 
 of the breast. The figure shows an appearance very much hke 
 that of an epithchum of many layers. The lowest layer appears 
 of a deeper colour, being made up of the more cylindrical and 
 pyramidal forms of cells closely packed together; and that is ]>re- 
 cisely the character of the lowest layer of a normal epithelium of 
 several layers, as, for example, that of the trachea. The upper 
 cells are rounder and separated by a larger amount of intercellular 
 substance, and the free border may be seen to run out into pro- 
 cesses of the same kind as those referred to in the preceding figure. 
 All the cells that line the walls of these acini, and in some ca>es 
 occupy the entire space of the acinus, are of the nuclear kind. 
 
 It may be observed that the columnar processes of the large 
 yellow cells, such as are figured on p. 131, showed a tendency to 
 bend over at their free ends. The same may be observed in Fig. 
 28. The round disc-like appearance in the upper of the two 
 
 Fig. 30. 
 
 rrom tlip same case as Fig. 29. An acinus filled with trabecular 
 processes of nuclear cells. Magnified 150 diameters. 
 
 processes, is evidently the transverse section of the process that has 
 doubled upon itself and is now growing out at an angle with its 
 original direction. This is very commonly observed ; and still 
 more commonly one finds that the space of the acinus is occupied 
 in all directions by trabecular columns of cells, which are nothing 
 else than the processes that have become fused together at their 
 free extremities. Fig. 30 is a drawing of an acinus from the case 
 that Fig. 29 has been taken from. The entire sj)aceof the acinus, 
 it will be observed, is occupied by anastomosing columns or 
 trabeculoe of cells ; the cells are all of the same nuclear kind, with 
 a certain amount of intercellular substance. The subjoined figure 
 
Cn. VI.] OF THE BREAST. U5 
 
 No. 31 has a great resemblance to the preceding, but it sliows 
 differences in the character of the component cells that are not 
 without sioniticauce. Fio-. 81 is from a tumour of the human 
 male breast. The patient, a man past middle age, had a tumour 
 in each breast, one of them quite small and firm, the other larger 
 and in parts softened. The drawing is taken from a firm and 
 undegenerated part, and represents the prevalent structure. The 
 growth consisted of greatly enlarged acini, sometimes packed closely 
 together, but in general with a certain amount of coarse fibrous 
 or chronic-inflammation tissue between them. Within the acini, 
 the epithelium was of a perfect polyhedric type, consisting of 
 
 Fig. 81. 
 
 
 
 
 
 From a tuuiour of the human male breast. An acinus filled with trabecular 
 columns of polyhedi'ic nucleated epithelium. Magnified 150 diameters. 
 
 nucleated cells with a broad fringe of almost hyaline protoplasm. 
 There was everywhere a considerable hyperplasia of the epithelial 
 cells, which sometimes gave the floor of an acinus a ridged appear- 
 ance, but more often resulted in the trabecular filling up of the free 
 space of the acinus, as shown in the figure. In some situations 
 the cells were more elongated or lozenge-shaped, but there was 
 everywhere the same relation between the central nucleus and the 
 investing cell-substance. When the cells were seen edgewise they 
 appeared to be thin plates, resembling squamous epithelium. The 
 only other point that requires to be mentioned by way of descrip- 
 tion is, that the softening or degeneration of one of the tumours con- 
 sisted in the caseation of large numbers of the detached epithelial 
 c. c. 10 
 
14G PATHOLOGICAL PROCESSES [Pt. TI. 
 
 cells within the acini. The appearance of the softened parts to 
 the naked eye was caseous, and with the microscope the acini were 
 seen to be filled with heaps of rounded cells, which had the usual 
 blurred outlines and opaque substance of cells that had become 
 caseous. 
 
 The significance of the form of the cells in this case lies in the 
 fact that it was a tumour of the male breast. The male breast, 
 although it developes for a time like the breast in the female, soon 
 loses the capacity of functional excitation with which it was originally 
 endowed ; while something of its structure remains. There is, as it 
 were, a severance of function from structure. But the structure is 
 still capable of excitation ; it may be subject to irritant influences 
 like any other part of the body. Tlie effect of the excitation, in 
 whatever way the latter operated, is seen to be a purely hyper- 
 plastic effect. The new growth of epithelium is polyhedric 
 nucleated epithelium, and not the crude products of a feeble 
 secretory stimulus. Again, the transformation of this epithe- 
 lium is not any of the transformations of secreting epithelium, 
 for instance into a mucous or a fatty substance, but it is a caseous 
 transformation^. 
 
 Continuing the comparison between the tw^o figures No. 80 and 
 No. ol, in the latter the hyperplastic process can be seen by itself; 
 if it is succeeded by another process in the individual cells, that 
 process is one of caseous degeneration. But in the diseased acinus 
 of the female breast, the hyperplastic or formative process and the 
 functional or transforming process are both implicitly present in 
 one and the same act. In tumours of the female breast the perfect 
 polyhedric kind of epithelium is seldom found ; the kind of cell 
 that is most commonly found both in skirrhous and medullary 
 cancers is the large nuclear cell. Measured on the physiological 
 
 ^ Caseous transformation of hyperplastic epithelium occurs elsewhere in the 
 hod}', very commonly in the epithelium of the lung alveoli and in the cells of the 
 suprarenal body, and it cannot escape notice that the cells in both of those situa- 
 tions are epithelial-like cells in exactly the same position as the epithelial cells of 
 the male breast; that is to say, they agree in being not subject to the ordinary 
 functional excitations of glandular epithelium, and they have a considerable 
 morphological resemblance. It may perhaps turn out to be a sound generalisation, 
 that the caseous transformation of hyperplastic epithelium befalls only those epi- 
 thelial cells which are capable of no secretory excitation. 
 
Ch. YI.] of the breast. U7 
 
 scale, those cells beloug to the intermediate stage of the breast's 
 uufolding, and they stand for a half-roused functional stimulus. 
 Although, in the cases from the bitch, there is good evidence that 
 a still feebler stimulus, bringing the function up from the resting 
 state, must have gone before, it is at the later or intermediate 
 stage that the pause in evolution is made, and it is at that level 
 that the continuous morbid action of the gland proceeds for an 
 indefinite time. 
 
 The effect of the continuous action of a feeble secretory force is 
 well seen in the Figures 28 to 30. Under ordinary circumstances, 
 the product that is formed within the acini, however crude it may 
 be, leaves the gland. There is abundant evidence that, in the 
 normal evolution, large numbers of the most imperfect secre- 
 tory products, viz. the pigmented cells, leave the gland by the 
 lymphatic channels or otherwise. Under the spurious evolutional 
 influence, they follow the same inherent tendency so far. But it 
 is not uncommon to find accumulations of them in situ, as shown 
 in tlie Figures 23 and 24. Although the morbid excitation 
 corresponds in its intensity to a stage of the normal evolution, 
 there is this fundamental difference, that the corresponding stage 
 of the normal process is transient, giving place to a stronger force, 
 while the morbid process continues indefinitely at the same en- 
 feebled level. The discharging force of the gland is therefore 
 weakened, inasmuch as the increment of the function is necessary 
 to its vigour. Elements that are really secretor}^ products of the 
 gland tend to accumulate within the acini, and the process takes on 
 the aspect of a purely formative or proliferative process. Nothing 
 could be more suggestive of proliferation than the almost vegeta- 
 tive growth of pigmented cells in Fig. 23 ; but nothing is more 
 certain, at the same time, than the functional significance of the 
 cells. They are themselves the crude secretion, and they differ 
 from their physiological prototypes only in accumulating within 
 the secreting structure. If the cells had less obvious marks of 
 functional transformation upon them, it might be said that the 
 distinction insisted upon between formative and functional action 
 is a merely theoretical distinction and without existence in fact. 
 But the characteristic pigmentation is an evidence that the cells do 
 not result from a formative impulse only ; it is evidence that the 
 dualism of function and structure is undissolved even in disease. 
 
 10—2 
 
U8 PATHOLOGICAL PROCESSES OF THE BPEAST. [Pt. II. 
 
 The case of the tumour of tlie male breast affords a convenient 
 illustrative contrast. In the male breast the function falls into 
 disuse at an early period, but something of the structure survives ; 
 the union of structure and function is as it were dissolved. Con- 
 sequently, the result of a morbid excitation might be expected 
 to be mere hyperplasia ; and so indeed it is in this case. The 
 form of the epithelium indicates hyperplasia alone ; and the only 
 change that it undergoes is in the way of caseous degeneration. 
 
 The cells that fill the acini in Figs. 28 — 30 are therefore not 
 only the epithelium of the acini, but they are at the same time 
 the crude products of the secretion. The formative process no 
 doubt outruns the functional, and that must be held to be one of 
 the most imjjortant points in the causation of th.e tumours. The 
 cell that should have passed out of the gland as a waste cell, 
 remains in the place of its origin and begins to multiply there. 
 The work of proliferation may be seen in Fig. 29. In the deepest 
 layer of cells, some of them are found to be elongated, with a 
 nucleolus at each end ; appearances of constriction in the middle 
 may be also observed, and the division of the cell into two may be 
 at once inferred. The cell that should have been thrown off from 
 the acinus almost as soon as it was formed, remains, with its pro- 
 geny, to infest the glandular structure. Such cellular products of 
 the function are inevitable in the breast, owing to its periodicity ; 
 the initial error consists in the gland being disturbed from its rest- 
 ing state by some unusual cause ; and the misfortune lies in the in- 
 definitely prolonged formation of crude cellular secretion, without 
 any approach to a higher and safer intensity of function, and in the 
 retention within the gland of the crude secretion, either as intra- 
 acinous accumulations, or as extra-acinous infiltrations. It is upon 
 such deviations from the physiological track that the existence of 
 a tumour depends. So far as relates to the large nuclear cells, the 
 intra-acinous collections of them correspond to the structure of 
 medullary cancer, and the extra-acinous infiltrations of the same 
 cells are a distinsfuishingf feature of skirrhus. 
 
CHAPTER VII. 
 
 PATHOLOGICAL PROCESSES OF THE BREAST-contlnned. 
 
 In attempting to refer to their phj-siological type the patho- 
 logical processes last discussed, I was under the disadvantage of 
 having for the normal standard a period of the evolution of the 
 breast that is itself difficult to explain. I pointed out in the 
 second chapter that the earliest evolutional changes in the breast 
 could be followed without difficulty, owing to the pigmentation of the 
 cellular products that characterise them; and that the latest changes 
 were intelligible, inasmuch as the vacuolated cells within the acini 
 differed from the colostrum or milk-containiug cells only in the 
 circumstance that their vacuoles were filled with mucus. The 
 large yellow granular cells are products of the secretion so well 
 marked in every way that their occurrence, both in healthy and in 
 morbid states, may be safely taken to support even the novel pa- 
 thological doctrine that has been stated. Hardly less reliable in 
 the pathological application are the phenomena of the breast in 
 the immature period which is characterised by the production of 
 mucus. A considerable proportion of the tumour cases in the 
 bitch have their physiological type in that condition of the organ 
 and in that intensity of the function. The physiological type is 
 that of the later period of the evolution, in which the function 
 comes near in its intensity to the perfect lactation. But there is 
 every reason to think that the morbid process, even when its 
 prevalent characters are those of advanced evolution, has started 
 from a resting condition of the breast, and has advanced in the 
 usual way through the earlier stages to the later. In the tumour 
 
150 PATHOLOGICAL PROCESSES [Pt. II. 
 
 cases about to bo described, the glandular tissue round the centres 
 of disease is for the most part in the resting state, and, in the more 
 diffused morbid enlargements of the gland, or in the younger 
 tumour nodules, one may find undoubted traces of the earlier stages 
 of evolution. In no other varieties of the tumour disease are the 
 pigmented cells so numerous, although the actual tumour has often 
 the myxomatous or cartilaginous character. 
 
 However far the spurious stimulation of disease may carry the 
 mamma from the resting state towards its perfect evolution and 
 its perfect secretory activity, the stages of the diseased process arc 
 always more tardy than in the evolution of pregnancy, and the 
 pathological investigation often affords an opportunity of detecting 
 some particular state of the mammary epithelium, which in the 
 normal study was rather a matter of logical proof than of actual 
 observation. Thus, in a diseased mamma, the pigmented granular 
 condition may be observed in the complete set of cells Avithin an 
 acinus, while in the evolution of health seldom more than two 
 or three pigmented cells are seen in situ, although they may be 
 numerous outside the secreting structures. The well-marked 
 stage of mucus production that occurs in the latter half of 
 evolution may be also more deliberately observed in the spurious 
 excitation of disease. The gland or some part of it becomes for 
 the time being a mucous gland. Fig. 32 shows two mammary 
 acini which are taken from different cases in the bitch. It 
 would be difficult to obtain from any part of the digestive tract 
 more perfect examples of mucus-producing structures. Although 
 there is abundant evidence that, in the healthy mamma, a 
 certain incomplete degree of the functional excitation both in 
 involution and in evolution has for its effect the production of 
 mucus from the secreting cells, the cells are not to be found ever 
 at that point of their existence when they have the characters of 
 perfect un transformed mucous cells. Those characters are, a some- 
 what columnar shape, the nucleus placed towards the attached 
 border of the cell, and a large amount of cell-substance which has 
 the peculiar and, as it were, succulent aspect found in secreting 
 cells belonging to that class. In the corresponding stage of the 
 evolution of pregnancy, the cells are found to be all of them 
 vacuolated more or less extensively, as in Fig. 17; the condition 
 of the gland is probably analogous to those states of the intestinal 
 
Ch. VII.] 
 
 OF THE BREAST. 
 
 151 
 
 mucous mcMubrano, or parts of it, where the cpitliclial cells are 
 foiiiKl in the condition of goblet cells. Why the production of 
 
 Fio. 32. 
 
 Mamrnary aciii', from two different patholofiipal cases in the bitch. The ejiithi Ihim 
 in the form of perfect mucus-producing columnar cells. Magnitied 3C0 diameters. 
 
 mucus from the epithelium of the mamma should coincide with a 
 subsiding or an incomplete degree of its functional force is not at 
 all clear ; but it is a well-established fact that the mucous trans- 
 formation of the mammary cells occurs at exactly corresponding 
 periods in the parallel processes of involution and evolution, and it 
 may be taken absolutely to be the characteristic effect of a limited 
 functional excitation. Corresponding to this degree of the secre- 
 tory force there is a group of pathological processes in the mamma 
 which are of frequent occurrence, and which are the point of de- 
 parture for many of the tumours in the bitch and for some of the 
 varieties of mammary cancer in the human subject. Such tu- 
 mours often correspond closely to the cystic sarcomas, or myxomas, 
 or colloid cancers ; but as we have been led, for sufficient reasons, 
 to avoid treating the tumours according to their classes, the names 
 are mentioned here solely to show what are the familiar tumours 
 that now ftiU to be discussed. 
 
 The mucous-gland structure of Fig. 82 is perfect of its kind, 
 but it is itself rare in the mamma, and is found more commonly 
 in certain states of transformation. Enlarged acini occur in which 
 the cells, forming a circlet round the wall, are spherical and 
 distended with mucus-like protoplasm, or actually vacuolated 
 like the cells in Fig. 8 on page 18. The cavities of acini are 
 not unfrequently seen to be filled with the same large vacuolated 
 cells. The distinction between them and the colostrum cells 
 proper is in their substance only; the form is the same, but 
 the substance, in the one, takes on the degree of staining cha- 
 racteristic of mucus or of mucus-yielding protoplasm, while in 
 the other it remains mostly unstained and of a greyish appearance. 
 
152 PATHOLOGICAL PROCESSES [Pt. II. 
 
 In some acini, the cells on their wall are the yellow cells of the 
 earliest evolutional stage, and there are transitional cases where 
 the pigmentation is fainter, while there is the same vacuolated 
 condition of the protoplasm as in the actual mucous cells. 
 
 These are the acini in which there is the smallest extent of 
 deviation from the normal. The transformations of the epithelium 
 that have the most direct relation to the tumour process are 
 somewhat different. For one of these varieties a brief description 
 must suffice. The acini are filled with large collections of mucus 
 penetrated with vesicles in all directions, and in this mucus lie 
 a number of round deeply-coloured nuclei. In favourable speci- 
 mens, the vesiculated mass of mucus is seen to be an aggregate 
 of many vacuolated cells which have the round nuclei in their 
 cavity or on their periphery. As to the cells that line the wall 
 (5f the acini in those cases, they are also nothing but deeply- 
 stained round nuclei, with no obvious cell-substance. Two cases 
 of tuinour in the bitch, one of which had a superficial cyst and 
 was highly malignant (the liver being full of secondary tumours), 
 were made up chiefly of groups of acini containing cells of the 
 above categoiy. Generally speaking, the acini were grouped in 
 the ordinary lobular arrangement ; so far, the change in the gland 
 occurred uniformly throughout its substance. The interlobular 
 tissue was extensive, juicy, and often filled with numbers of the 
 large yellow and other cells. Inside the acini the cells were either 
 in more layers than one, or they ran out into processes, or they 
 completely filled up the acinous space. In these points the patho- 
 logical process resembled that already described for two other 
 varieties of waste products, the large yellow cells,, and the kind 
 of cells in Fig. 28 and Fig. 29 ; the intra-acinous papillary pro- 
 cesses seemed to begin with a submucous collection of the nuclear 
 cells, which had the effect of raising the surface-cells. The chief 
 difference lies in the fact that the cells now under notice are 
 smaller and rounder, or more perfectly nuclear. Where the vacuo- 
 lation process is obvious, they are the correlated cellular product 
 of a more complete vacuolation. Where the vacuolation is 
 not so obvious, they correspond to the Avaste cells where the 
 functional excitation is considerable, and they have the closest 
 resemblance to those drawn in Fig. 9, and in Figs. 5 and G. 
 
 The normal mamma (of involution), from which the two last 
 
Cn. VII.] OF THE BREAST. 153 
 
 are taken, shows other celkilar products of vacuolation besides 
 the small round cells. These are the cresceutic, or rhombic, or 
 triangular, or rod-shaped cellular bodies, which are derived from 
 the vacuolation of cells in a perfectly intelligible manner. The 
 same varieties of shape are found among the cells within the 
 acini in the pathological states of the mamma that correspond 
 to the mucus-producing period. In particular, the greater part 
 of the tumour in another of the cases in the bitch consists of 
 alveolar groups of these cells, and the alveolar arrangement is 
 clearly enough seen to correspond with the earlier acinous arrange- 
 ment. The cells, while retaining their various shapes, have very 
 generally a narrow fringe of protoplasm round the deeply-coloured 
 nucleus. This tumour might be considered a myxoma. 
 
 There is, however, among the cases collected, one in particular 
 ■which illustrates so many of the pathological processes at once, 
 that it may be selected for a more detailed and circumstantial 
 description. The animal was an old setter bitch. In July 1873, 
 a tumour was removed from one part of the mammary chain, 
 globular in shape and lobulated, about two inches in diameter, 
 firm in consistence and yielding on section a clear yellowish or 
 brownish juice. There was one very small point of cartilaginous 
 tissue near its centre. This tumour belonged to a variety that 
 will fall to be described afterwards, among the concluding illus- 
 trations of pathological processes. In June 1875 the animal was 
 brought back for treatment, the disease having re-appeared. 
 There was now a large soft tumour towards the inguinal end of 
 the other mammary chain, and there were several smaller nodules 
 throughout the whole length of the mammce on both sides, besides 
 a more diffused enlargement about the middle of the gland on 
 one side. The animal was considered to be past surgical treat- 
 ment and Avas killed. The liver contained a number of small 
 rudimentary nodules, and in the mediastinum there was an 
 enormous lobulated mass of medullary consistence. At a point 
 in the vena cava inferior about two inches below its termination 
 in the auricle there was an oval mass of the same consistence, 
 rather smaller than a hen's egg; it appeared to be connected 
 with the tumour-mass in the mediastinum. 
 
 Some of the small mammary nodules were cartilaginous; others 
 did not differ in crross characters from the aland-tissue that 
 
154 PATHOLOGICAL PROCESSES [Ft. II. 
 
 surrounded them except in being circumscribed and of greater 
 density or closeness of texture. The diffused enlargement, again, 
 which extended over two inches of the mammary chain, obviously 
 owed its special character to some process going on more or 
 less uniformly throughout the gland over that area. To take 
 advantage of the evident richness of this material, several sets 
 of sections were made from different portions of it. Tvlany parts 
 of the mammary chain of glands were found to shoAv the perfectly 
 typical resting state. Another common appearance was that of 
 acini in the first evolution stage, with large yellow cells within 
 and around them. Some of the smaller nodules were made 
 up in part of collections of the pigmented cells in the fibrillar 
 tissue. Sometimes an entire lobulus is encountered, the acini 
 of which have the mosaic of perfect epithelium as in the fully 
 expanded gland. The epithelial cells, however, have unusually 
 large and oblong nuclei, as shown in the right-hand group of 
 cells in Fig. 2 drawn from another case. But the processes most 
 characteristic of the case belong to the stage of mucus production. 
 One of the acini of Fig. 31, showing perfect mucous cells, is taken 
 from this case. Other acini, again, have a circlet of spherical 
 mucous cells, or the same vacuolated ; and there are to be seeil 
 in reality those gradations in tint between the distended spherical 
 mucous cells and the large yellow cells of an earlier stage, which, 
 in the study of the physiological evolution, are rather an inference 
 than a matter of observation. 
 
 But the most instructive appearance of cells within the acini 
 is that drawn in Fig. .33. The figure contains two adjoining acini, 
 the upper of which is incomplete, but shows very well the cells 
 individually. The lower is a complete acinus with two apparent 
 diverticula, which are no doubt the infundibula of adjacent acini 
 that have been included in the same plane of section. The acinus, 
 it will be seen, has a small free space in the centre, and several 
 layers of cells on its wall. This is the same pathological process 
 that we have had to consider before, and as in the former cases, 
 except in that of the male breast, the many-layered epithelium 
 implies more than a mere hyperplasia. In the cancer of the male 
 breast the cells were perfect epithelial cells, but in the other cases 
 of hyperplastic acini, as in this, the cells are of another type. In 
 Fig. 29, from a soft cancer of the female breast, the lowest layers 
 
Cn. YIL] 
 
 OF THE BREAST. 
 
 155 
 
 are oblong nuclear cells, while the upper cells are rounder. In the 
 present case, there is the same admixture of elongated nuclear and 
 other cells. In both cases the form of the cells is owing to the 
 fact that there is not merely a hyperplastic process going on, but 
 a functional process as well ; there is in fact the dual activity of 
 the secreting gland. In the present case, the activity corresponds 
 to that degree of excitation which produces mucus from the epi- 
 thelium. One acinus from this case has been figured already in 
 which the epithelium has the perfect mucous type, and the same 
 kind of cylindrical cell with the nucleus at its base may be seen in 
 the half acinus at the top of Fig. 33. The obvious and uumis- 
 
 FlG. 33. 
 
 Mammary acini of the bitch (pathological). The epithelium massed in 
 several la.vers; the cells columnar and nucleated, or spindle-shaped. 
 Magnified SOO diameters. 
 
 takeable secretory change in the cell is where it becomes globular 
 and vacuolates, and we have frequently had occasion to notice, 
 both in the healthy processes and in the morbid, that the vacuo- 
 lation may be associated with the production of a waste cell. 
 But, where we find the formative excitation outrunning the 
 functional, and the cells produced relatively so fast that they 
 accumulate in papillary processes or in more layers than one, the 
 secretory change is not so obvious. It is shown only in the altered 
 shape and proportions of the cell. In all such cases of hyperplasia 
 
156 PATIIOLOOICAL PliOC ESSES [Pt. II. 
 
 belonging to the stage or degree of mucus production, the secreting 
 cell tends to become an oblong, staff-shaped, or spindle-shaped cell. 
 Among these pathological processes in the mamma, a whole lobulus 
 is sometimes found in which the acini are lined with long, narrow, 
 staff-shaped nuclear bodies^ In the figure now being described, it 
 will be seen that they are mostly spindle-shaped. In the lowest 
 layer a large pyramid cell occurs, with three nuclei. 
 
 The accumulations of epitlielium within such acini as are re- 
 presented in the last figure stand for more than a mere hyper- 
 plasia; the transforming or functional force is also present, and its 
 effect is shown in the forms of the cells. But the presence of the 
 functional factor is also shown in a much less doubtful manner. 
 As the yellow cells of an earlier stage could be seen, both in nor- 
 mal and morbid cases, not only within the acini but also in the 
 interacinous and interlobular tissue and ultimately in the lymph 
 sinuses of th3 lymphatic glands; so the mucus-yielding epithelial 
 cells of the present stage may be seen, in one and the same pre- 
 paration, occupying the position of epithelial cells, and collected 
 in heaps in the interacinous and interlobular tissue. Reverting to 
 Fig. 82, in which the mucus-yielding cells are of a- somewdiat 
 cubical or columnar shape, cells of precisely the same form and 
 substance are found in rows just outside the acini. Were it not 
 for the analogy of the large yellow cells, whose escape from the 
 acini into the neighbouring tissue cannot be questioned, the oc- 
 currence of large cubical epithelial cells outside the acini might 
 appear to depend on some error of interpretation. It might, for 
 example, be urged that the groups of cells that appeared to be 
 extra-acinous were merely the contents of acini whose walls had 
 given way in the section. But it is not more surprising that the 
 mucus-yielding cells should leave the acini bodily, than that the 
 pigmented cells of the more immature type should do the same. 
 In the same preparations there are often to be seen circular belts 
 of the yellow cells occupying the interacinous tissue, and near 
 them similar interacinous accumulations of non-pigmented cells, 
 sometimes of the smaller nuclear type, but not unfrequently of the 
 more characteristic mucous type. The most common form of the 
 latter is not the cubical or cylindrical cell, but a spindle-shaped 
 
 1 An aiijiearauco of tliis kiud is ligured by Laughaus, VircJiow's Archie, Vol. 58, 
 Plate 3, Fig. 4. 
 
Oil. VII.] OF THE BREAST. 157 
 
 cell of the kind that occurs most numerously in Fig. 83. The 
 difficulty that one has to contend against here, one may say the 
 incubus that has here to be shaken off, is the theory of the connec- 
 tive-tissue origin of all such cells. But two facts are established : 
 first, that crude forms of epithelium do escape from the acini into 
 the surrounding connective tissue, aud second, that one of the forms 
 of the crude epithelial cells is the spindle-shaped cell (Fig. 33). 
 However apt, therefore, one may be to take the extra-acinous 
 collections of spindle cells for products of the connective tissue in 
 which they are found, that simple and conventional explanation 
 must give way to one that is more in accordance with the facts. 
 Cells of the oblong or spindle form are found as a somewhat mor- 
 bid type of epithelium, and it is nothing more than their destiny, 
 according to the normal scheme of the function, that they should 
 leave the acini. 
 
 In the particular case which has been chosen to illustrate this 
 variety of the tumour disease, the collections of fusiform or oblong 
 cells outside the acini are very numerous. In the part of the 
 glandular chain where the process is more uniformly diffused, 
 they occur side by side with the other kinds of waste products. 
 The smaller tumour nodules are in great part made up of ag- 
 gregates of the elongated and spindle-shaped cells between the 
 acini ; it is their presence or their retention there that changes the 
 glandular structure at various points into a corresponding num- 
 ber of small tumours. The more extensive the interacinous col- 
 lections of waste cells are, the more do the outlines of the acini be- 
 come obscured. In this manner the structure of the diseased 
 breast comes to be that of an " adeno-sarcoma," the sarcomatous 
 element being represented by the spindle or other cells outside the 
 secreting structure, and the adenomatous element by the same 
 kind of epithelial cells that are still in situ. Some of the less 
 cystic varieties of " cystic sarcoma " correspond to the same morbid 
 condition of the breast. 
 
 The particular case that has been so far described contained 
 one very large tumour and a number of smaller nodules. It now 
 remains to give an account of the former, and the explanation of 
 the smaller nodules will be found to lead up to the theory of the 
 principal growth. The large tumour occurred towards the inguinal 
 end of one of the glands; it was globular in shape, about three 
 
158 PATHOLOGICAL PROCESSES. [Pt. II. 
 
 inclics in diameter, of uniform soft consistence and whitish colour. 
 Fig. 34 represents the minute structure as it appeared in sections 
 taken from a central and representative part of the mass. The pre- 
 dominant cells are oblong or fusiform, and of granular substance. In 
 every part of the section there is seen a certain arrangement of the 
 cells which is not sufficiently conveyed by the figure. The tracts 
 of oblong or fusiform cells follow a sort of figure-of-eiglit course ; 
 they enclose somewhat lighter or more empty spaces among them 
 
 PiCt. 31 
 
 
 k 
 
 
 
 
 hi 
 
 JAf'Mill/y y^t^i 
 
 
 
 From a mammary tumour in tlie bitcli. Broad tracts of oliloiip 
 cells eiiclosinf; alveolar spaces wliicli contain other (onus of cells 
 and mucus. SlaKnilied SUO diameters. 
 
 at regular intervals. The lighter spaces are occupied by rounder 
 and often larger cells, and by intercellular substance of a mucous 
 nature. The tumour, as an entity 'or individual, might be 
 classed among the myxo-sarcomata, but, according to the rational- 
 istic interpretation of its growth, it represents a condition of the 
 structure of the breast in which the extra-acinous collections of 
 cells have encroached so much upon the acini as almost to ob- 
 literate them. The lighter alveolar spaces that occur at regidar 
 intervals throughout the section are the original acinous cavities, 
 
Cn. VII.] OF rilE BREAST. 159 
 
 occupied by a more spherical variety of cells and by mucus. The 
 structure is a more extreme form of the extra-acinous retention of 
 waste epithelial products that may be seen, in all stages and de- 
 grees, in the other and earlier centres of morbid action in the same 
 mamma. In the largest tumour there is no doubt evidence of 
 rapid multiplication of the oblong and fusiform cells in their extra- 
 acinous situation ; but there is no evidence whatever that the 
 proliferating cells were derived originally from the connective- 
 tissue corpuscles of the stroma of the organ. I am well aware 
 that this account of the origin of the tumour is in direct opposition 
 to an opinion that is universally held. But I submit, at the same 
 time, that it so differs because it takes into consideration a factor 
 in tumour-formation that is all but universally ignored, namely, 
 the physiological or functional factor. Whether in health or in 
 disease, there is an indissoluble union between the mammary 
 structure and the mammary function, and that dualism is nowhere 
 more apparent than in those immature periods of the breast which 
 precede the perfect lactation. If the entire range of processes in 
 the breast be set up as the, physiological standard, such tumours as 
 the one last described will not only seem to be the natural outcome 
 of functional irregularity, but, on the other hand, the constantly 
 assumed activity of the connective-tissue stroma will appear, in 
 the light of that physiological knowledge, to be arbitrary as an 
 assumption, and insufficient as an explanation. 
 
 The concluding illustrations of morbid processes, that are 
 furnished by the series of cases from the bitch, relate to tumours 
 in which there is a tendency to the formation of a kind of tissue 
 that is cartilaginous in its hardness and sometimes resembles 
 hyaline cartilage in its minute structure. The cartilaginous tissue 
 in the breast appears in no case to develope as cartilage from 
 the outset; it is always a cartilaginous transformation of some 
 earlier condition of tissue, and generally of myxomatous tissue. 
 The gristly character of the tissue depends upon a large amount 
 • of intercellular substance, which has become hard and firm like 
 the intercellular substance of hyaline cartilage. The cartilaginous 
 transformation is very apt to occur in mammary tumours of the 
 bitch, and it is more often seen at limited points of the tumour 
 than as a uniform change of its whole substance. A case occurred 
 of an extensive tumour in the bitch, which was mainly soft 
 
IGO PATHOLOGICAL PROCESSES. [Pt. IT. 
 
 and succulent, but at certain places hard or gristly, and, at one 
 central spot, truly cartilaginous and even calcified. There were 
 numerous secondary tumours in the lungs, which reproduced the 
 structure of the soft and succulent parts in the primary tumour. 
 There is every reason to think that the cartilaginous transforma- 
 tion is a safe and, as it were, cicatricial issue of the tumour 
 process, and that the infectiveness of the tumour diminishes in 
 proportion as the sclerosis of its substance spreads*. 
 
 In approaching the consideration of tumours of the breast 
 that contain cartilage, or that are on the way to become cartilage 
 wholly or partially, it might appear as if the physiological method, 
 which has been sedulously followed in the preceding sections of 
 the pathological investigation, must at length be given up. Is 
 it conceivable that the epithelial structure and the secretory force 
 of the breast should be in any way concerned in the production 
 of a cartilaginous tissue ? That, however, will be the conclusion 
 drawn from the following descriptions and illustrations ; and as 
 the conclusion is an unusual one, I must premise that the subject 
 is introduced here not for the sake of putting forward quixotic 
 theories, but simply in order to complete the series of morbid 
 changes which this collection of mammary tumours exemplifies. If 
 it be personally a more agreeable task to substantiate the current 
 opinions, it would no doubt also be prudent in the present case 
 to abstain from adding still further to the unaccredited pathological 
 doctrines that I have been endeavouring to expound and to 
 establish. The cartilaginous forms of tumours might be disposed 
 of, in a sentence, as growths of the connective tissue, and as 
 unconcerned, therefore, in the physiological processes of the gland. 
 But a prolonged study of mammary tumours in the bitch will 
 satisfy anyone that there is no line of separation between the 
 cartilaginous and the other forms of tumours. The assumption 
 of their origin from the connective tissue of the gland might 
 be easily made, and would be readily admitted. But such an 
 assumption would be dictated by indolence, and I shall now en- 
 deavour to show that it would be based on error. 
 
 ^ The ordiuary tumours of the parotid, partly myxomatous aud partly cartila- 
 ginous, have a general resemblance to the mammary tumours in the hitch, referred 
 to in the text. The parotid tumours are known to he comparatively innocent 
 growths, which may have existed for many years without infecting the system. 
 
Cn. VII. ] 
 
 OF THE BREAST. 
 
 IGl 
 
 Fig. 35 is a drawing of several contiguous acini from a case 
 that had the cartilaginous tendency in a very marked degree. 
 The large principal tumour was partly cartilaginous as well as 
 partly osseous, aud the smaller nodules exemplified other processes 
 as well. The fringes of the more or less normal mamma showed 
 exquisitely the process of pigment-cell formation in the acini, 
 
 Fig. 35. 
 
 ^^ML^f 
 
 From a mammary tuiti'iur in Vac bitch. A a;roun of four acini, showing various 
 changes of tlie epithelium. Blagniiied :;10 diameters. 
 
 and many parts of the tumours had immense numbers of the 
 large yellow cells in the fibrillar tissue. But the pathological 
 processes, as a whole, have their type in another part of the 
 physiological scheme. One of the acini in the figure will be 
 found to have a trabecular growth within it not unlike those 
 already described and figured ; another shows a papillary process ; 
 while the whole of them show an unusual condition of their 
 epithelial lining. The marginal row of cells are round nuclei of 
 c. c. 11 
 
1G2 PATHOLOGICAL PROCESSES [Pt. II. 
 
 considerable size embedded in some intercellular substance. In 
 that respect the condition is the same as in the other cases of 
 man^'^-layered hyperplastic acini. But the condition in the lower 
 la3'ers is somewhat special, and it is in the differences exemplified 
 there that the distinctive characters of this cartilaginous process 
 consist. That difference may be said in a word to be the more 
 real vacuolation of the cells, corresponding to a greater intensity 
 of the functional stimulus. 
 
 In treating of the pathological processes to which the tumour 
 of Figs. 32 — 34 is to be referred, it was mentioned that the perfect 
 mucous cells were elongated cylinders, that an entire lobulus was 
 sometimes found in which the epithelium was in tiie form of long 
 narrow rod- shaped cells, and that the elongated or spindle form of 
 the epithelium was especially well marked where there were more 
 layers than one. The perfect mucous cells, as drawn in the pair 
 of acini already mentioned, had a nucleus at their attached end 
 and a large amount of granular cell-substance, but the cells of 
 the pathological acinus were filiform or rod-shaped, or spindle- 
 shaped. That circumstance was taken to indicate that the cells 
 had undergone a certain degree of functional transformation ; that 
 they were not perfect epithelial cells, but in reality crude agents or 
 products of the secretion. Whatever secretion proper they had 
 given origin to was in the form of intercellular substance, but 
 otherwise it might be said that the whole epithelial cell became 
 waste. Now, the position that has to be advanced at this jjoint 
 of the investigation is, that a very common class of slender rod- 
 shaped or crescentic bodies found in the particular pathological 
 processes at present under review, are phenomena of the same 
 class. The acini in Fig. 35 show several of these. In many cases 
 the deeply-coloured crescentic line evidently corresponds to one 
 side of a cell or of a nucleus ; it represents in fact the peripheral 
 solid part of a cell undergoing some kind of vacuolar trans- 
 formation. The peculiarity of the pathological process is that 
 the vacuolation does not result in the production of an actual 
 fluid which is discharged, but of a hyaline intercellular substance, 
 among which the deeply-coloured straight or curved bodies, as well 
 as paler round nuclei lie. This account of the process will become 
 much more intelligible and will doubtless gain in credibility by 
 a reference to Fig. 30, which is taken from the same case as the 
 
Ch. VII.] 
 
 OF THE BREAST. 
 
 1G3 
 
 preceding. The acini contain a number of large cells resembling 
 fat-cells. But the appearance in the preparation, although the 
 woodcut may fail to give it so completely, is not that of cells 
 filled with fluid, but of cells filled with a firm hyaline substance. 
 The cells are in fact the vacuolated epithelium ; the hyaline 
 substance corresponds to the contents of vacuoles, and the cres- 
 ceutic or rib-shaped bodies to the peripheral masses. The figure 
 
 Fig. 36. 
 
 From the same case of mammary tumour as Fip. 35. A group of a?ini distinguished 
 by their hyaline appearance. Extensive vacuolatioii of the epithelial cells: hyaline 
 contents of the vacuoles. Magnilied 240 diameters. 
 
 miorht be matched with another almost similar from a different 
 case, and in the latter there may be seen in the immediate 
 neighbourhood of the hyaline-looking acini, an acinus in which the 
 cells round the wall have the ordinary appearance of vacuolated 
 cells, with peripheral nuclei of considerable size. It is quite 
 exceptional, however, to find the thin crescent-shaped or fibre- 
 like bodies in their original relation to a vacuolated cell, and 
 there is rea.son to think that the functional change to which they 
 
 11—2 
 
1G4 PATHOLOGICAL PROCESSES [Pt. II. 
 
 are due is often not an obvious vacuolation at all. They are 
 generally found scattered among round nuclear cells and in a 
 hyaline matrix. Isolated in this manner, they are apt to be 
 mistaken for the fibres of the connective-tissue framework in 
 which epithelial cells are supposed to be set. But besides their 
 relation to vacuolated cells, there is another position in which 
 they are found that may clear up any doubts as to their real 
 nature. One of the acini in Fig. 35 shows a trabecular band 
 stretching across it, and the appearance is not uncommon in this 
 and other cases. Sometimes the trabecules are much narrower 
 than that in the figure ; the free space of the acinus is bridged 
 over by a sort of network. These trabeculse at once suggest the 
 epithelial trabeculoe already described in two cases of cancer of 
 the human subject. In the present case they are made up of 
 long rod-shaped cells and a quantity of hyaline intercellular sub- 
 stance ; the principal difference between them and the trabeculae 
 of the other cases is that their cells are longer, narrower, and 
 almost like fibres ; but that is a difference that is quite in keeping 
 with the altered circumstances of the case. 
 
 The two illustrations last introduced have the great advantage 
 as regards brevity and directness, in that they show the patho- 
 logical process within acini that are simply enlarged ; the 
 new growth so far depends on collections of cells in the lumen of 
 acini, on several layers of cells round the wall, or on trabecular 
 bands stretching across or filling up the space. But although 
 the peculiar transformation of the individual cells is every- 
 where the same, that change in the cell is not, as a rule, asso- 
 ciated with mere enlargement of the acini. Perhaps the most 
 characteristic feature of these tumours is the growth in the acini 
 (and in the ducts as well) of papillary and bud-like processes, 
 which sometimes reach an enormous size. The bud-like growths 
 are attached to the wall of the acinus by a short pedicle ; the 
 expansion of the process is globular, and sometimes its free ex- 
 tremity bends over, and, as it were, grows downwards again. The 
 dilatation of a single acinus by such an excrescence is sometimes 
 very great. The free border of the growth is lined by a row of the 
 round nuclear cells separated by some intercellular substance ; the 
 appearance is the same as a marginal row of cells where there 
 are several layers in the acinus, and the appearance in the centre 
 
Cii. VII.] OF THE BREAST. 1(55 
 
 of the bud-like prominence corresponds also to that of the lower 
 stratum of cells. Where the process is well advanced, the sub- 
 stance of the intra-aciuous or intra-canalicular papillary growth is 
 made up of a hyaline matrix with a number of round nuclei and 
 many of the fibre-like or crescentic bodies embedded in it. But 
 papillary processes may be seen which are almost entirely cellular, 
 with some intercellular substance, and otliers, again, seem to be 
 made up almost entirely of a mucus-like homogeneous substance 
 which stains deeply. Another kind of papillary process has an 
 arborescent structure runninsf rather to the length than the breadth, 
 and these are the cases where the filiform or rod-like cells are 
 most abundant. How those various structures sometimes become 
 true cartilage is not easy to make out, although the fact is undoubted. 
 If a portion of the tumour that has the naked-eye character of 
 cartilage be examined, it will often be found that the cellular 
 elements are quite indistinct in the midst of the hyaline matrix, 
 and the common appearance is that of the crescent or rib-shaped 
 dark-coloured bodies forming one side of what obviously corresponds 
 to the future cartilage capsule. The tissue is, in that respect, not 
 unlike the peculiar cartilage that is found in callus. Finally, there 
 is an occasional development of the cartilaginous tissue into osseous 
 tissue. Some preparations show perfect examples of medullary 
 spaces lined with osteoblasts, and trabecular containing bone cor- 
 puscles. These appearances are so characteristic that I have oc- 
 casionally exhibited them as preparations of the normal osteo- 
 blastic condition of growing bone. 
 
 The bud-like or polypous growths that are so often associated 
 with the myxomatous and cartilaginous transformations are struc- 
 tures that may seem, in their mature form, to bear out the theory 
 of a connective-tissue origin ; they correspond to the kind of tu- 
 mour described as intra-canalicular polypous myxoma, and the 
 dendriform outgrowths, made up largely of rod-like cells, correspond 
 to the intra-canalicular papillary fibromata. But the same trans- 
 formation is represented in Figs. 3 ) and 36 as occurring throughout 
 groups of acini, and there is no doubt that the ordinary cellular 
 contents of the acini are the elements that become the subject of 
 it. The cells in the deeper layers of Fig, 85 are exactly the same 
 as occur in the substance of the bud-like growths, and the deeper 
 cells in the former are to be reckoned among the epithelial cells of 
 
166 PATHOLOGICAL ritOC ESSES [Pt. TI. 
 
 the acinus just as much as the more uniform rows of cells on the 
 surface. In the upper acinus of Fig. 85 there is, on one side, a 
 papilkiry outgrowth of cells, which may be regarded as an early 
 condition of the larger polypous growths that are seen elscAvhere ; 
 and that papillary outgrowth is obviously of the same kind as the 
 wart-like growths of epithelium in Fig. 28, If the epithelium of 
 the breast were at all times the polyhedric nucleated cells of the 
 lactation period, it would be somewhat irrational to derive a certain 
 class of the morbid structures from it ; but if it be borne in mind 
 that the epitiielium of the breast, in the course even of its normal 
 periodical processes, assumes a variety of types from stage to stage, 
 and that in its abnormal processes that variety takes a still wider 
 range, it will not appear strange, but on the other hand perfectly 
 reasonable to identify these varieties in the tumours that seem at 
 first sight to have grown from tlie stroma of the organ. On a 
 priori grounds it might be objected that epithelial cells and cells 
 of the connective series are not "equivalents;" but even as an 
 a priori argument, that objection is of (piestionable force. In my 
 paper on the histogenesis of secondary tumours in the liver, printed 
 elsewhere, I stated a number of facts to prove that secondary 
 sarcomatous nodules, no less than secondary epithelial nodules, 
 were formed out of the liver cells at particular points in the organ. 
 Thus the liver-nodules that were secondary to a sarcoma of large 
 and pigmented spindle-cells, primary in the subcutaneous tissue, 
 took origin in a tr<insformation of the liver cells at certain centres 
 of infection ; the individual liver cells could be seen to undergo a 
 definite endogenous change, so as to become the ultimate spindle- 
 cells of the tumour. In that case there is no equivalency as 
 between cells of the same type, but there is metaplasia or change 
 of type ; and if it be necessary to add a theoretical explanation to 
 the actual description of the myxomatous and cartilaginous growths, 
 that point of theory is supplied by the doctrine of metaplasia 
 which I have already stated in connexion with other varieties of 
 growths in the breast. 
 
 With the cartilaginous and myxomatous transformations, I 
 bring to an end the illustrations of morbid physiological action in 
 the mamma of the bitch. I have described, in their j)hysiological 
 order, the processes of tumour- formation in which the large pig- 
 mented cells, the somewhat larne and granular nuclear cells, and 
 
Ch. YIL] OF rilE BREAST. 107 
 
 the miicous cells (with their myxomatous and cartilaginous de- 
 velopments) respectively play a part. The illustrations that have 
 been given relate to the epithelial products both in the tissues 
 outside the acini, and in their original intra-acino\is position. The 
 intra-acinous accumulations are of very much the same nature in 
 all varieties of the tumour process ; the acini may have more than 
 one layer of cells on their wall, or their cavity may be occupied by 
 cells wholly or partially, or the epithelial lining may run out into 
 villous or papillary processes, or the acinous space may be crossed 
 by trabecular columns of cells. 
 
 In making use of the considerable series of cases in the bitch that 
 had been collected as the material of an investigation, two courses 
 were open to follow. Either the tumours might have been fully 
 described one by one with all their attendant circumstances, and, 
 as a final solution of them, they might have been referred to their 
 classes in the tumour-kingdom, after the manner of living things 
 in systematic zoology or botany ; or they might have been used 
 collectively to illustrate the various points of tumour-formation. 
 It is the latter course that has been followed, and it may be stated 
 with some confidence that the illustrations of pathological pro- 
 cesses, which have been given in their physiological order, do not 
 omit the leading features of any one of the whole series of tumours. 
 On the other hand, the mammary tumours of the bitch do not lend 
 themselves readily to the classification method ; they are some- 
 times multiple in the same case, with varieties of structure in the 
 several centres of disease ; and even when there is no multiplicity 
 of centres, the structure throughout the tumour is not always 
 uniform. 
 
 As the tumour disease in the bitch differs in some respects from 
 the mammary tumours of ordinary surgical practice, I collected at 
 the same time a number of cancers of the human female breast, 
 to which I shall now briefly refer for comparison. The figures 29 
 and 30 are taken from a very typical case of cancer in the breast. 
 The patient was a widow, aged GO, who had borne a large family. 
 The tumour was a circumscribed nodule, not more than one inch 
 in diameter, and situated towards the periphery of the gland on 
 the axillary side. The nodule was of firm texture, and it was 
 easily distinguishable from the surrounding tissue by the softer 
 and more parenchymatous character of its structure. According to 
 
1G8 PATHOLOGICAL PROCESSES [Pt. II. 
 
 the miscroscopic examination^ it must be considered to be an 
 intra-acinous formation occurring in a small and isolated portion 
 of the gland. The whole breast was removed in the operation, 
 and the bulk of it, which remained hard and dense after lying in 
 spirit, was made up of close tendon-like bundles of fibrous tissue, 
 with very few traces of epithelial tubes scattered through it at wide 
 intervals. Both in the density of its stroma and in the reduction 
 of its secretory framework, the tissue beside the tumour may be 
 considered to represent the usual condition of the breast in women 
 after the climacteric period. It is at that period of life that the 
 intensely hard tumours of the breast are apt to occur, and the 
 peculiar hardness is probably owing to the nature of the stroma in 
 which the cells, or active elements of the tumour, lie\ The pre- 
 sent case is an instance of a mammary tumour appearing late in 
 life, but occurring in the form of greatly enlarged acini of the 
 gland. In no other part of the organ was there any appearance 
 of a surviving cluster of acini, or even of the infundibula of ducts. 
 The glandular apparatus had disappeared, except at the point 
 where the morbid action had established itself. I have already 
 referred to the climacteric changes of the breast, although I have 
 not had an opportunity of following them closely ; and I observed 
 that the post-climacteric tumours of the mamma were to be con- 
 sidered as morbid growths incidental to the process of efFacement 
 of the secreting structure aud to the final withdrawal of the secre- 
 tory force. From that point of view, the origin of the present 
 tumour would be looked for in the survival of some outlying 
 part of the structure, and the correlated persistence of a func- 
 tional force, Avhile all the rest of the organ had become obsolete. 
 
 Fig. 26 is from another case in a woman beyond the climacteric 
 period, the patient's age being 58 ; it was removed on the same 
 day as the tumour just mentioned, and it offers several points of 
 contrast. The growth was more centrally disposed beneath the 
 nipple, and had not so definite boundaries. A considerable pro- 
 
 ^ "Thus we are told that scirrhus is marked and distingished by its fibrous 
 character : whereas the fibrous aijpearaiice, iu a great degree, belongs to the normal 
 condition of the organ, and is not a product of the disease. Many other equally 
 erroneous statements might be mentioned, and for this reason I felt that it was 
 absolutely necessary to give an account of the natural structure of the breast, before 
 its morbid changes could be properly explained or understood." — Astley Cooper, 
 Introduction to Anatomy of the Breast. 
 
Ch. VII.] OF THE BREAST. 169 
 
 portion of the cells appeared to fill up the surviving tubular 
 framework of the glaud, changing it into a system of solid tracts of 
 cells ; but a large number of the cells were as if infiltrated into the 
 surrounding stroma, in the manner shown in the figure. The 
 tumour was intensely hard, and may be considered to be a typical 
 case of skirrhus, 
 
 A third case, belonging to the same period of life, occurred in 
 a woman aged 56. The new growth was of small extent and 
 situated under the nipple ; it was indefinitely bounded, and the 
 stroma of the gland, or at least the substance round the tumour, 
 was chiefly fat-tissue which extended into the tumour area in the 
 form of pointed processes. Unlike the last-mentioned case of 
 post-climacteric cancer, this case showed several very distinct 
 acini, which were enlarged and occupied by anastomosing trabe- 
 culse of somewhat nuclear cells. 
 
 A fourth case occurred in a woman aged 54. The ducts of the 
 gland were obviously dilated, and a fluid of greyish colour, like a 
 catarrhal secretion, escaped from them on section. There were in 
 this case none of the usual appearances of an active growth, but, 
 on the other hand, evidences of retrogression or cicatrization. The 
 connective tissue was generally full of small round cells, and 
 resembled the tissue of chronic inflammation. At a number of 
 points of the glandular tissue, there were found concentrically 
 laminated sand-grains, which coloured deep purple with the log- 
 wood staining fluid. I have found the same kind of sand-grains 
 in one of the tumour cases in the bitch, and also in a ewe's udder, 
 in which there had been a somewhat disturbed periodical evolu- 
 tion. In the latter case, the sand-grains lay within acini of the 
 gland, and completely filled the acinous space. The occurrence of 
 sand-grains in tumours of the breast has been described as 
 evidence of a retrogressive process^; the cicatricial condition of the 
 connective tissue in this case (small-celled infiltration) is also 
 evidence of the innocuous nature of the disorder ; and I have to 
 add a third feature pointing to the same conclusion. In the 
 midst of the cicatricial connective-tissue there occurred, at some- 
 what regular intervals, groups of cells of a totally different kind ; 
 they were as large as the full-sized epithelial cells of the breast, 
 
 1 See a case of Carcinoma atrophicum of the breast, recorded by Ackermann iu 
 the 45th volume of Virchow's Archiv. 
 
170 PATHOLOGICAL PROCESSES [Pt. II. 
 
 and the clusters of tliem no doubt represented the remains of the 
 secreting structure. The point of interest is tliat each cell was 
 very distinctly vacuolated, having a nucleus of considerable size on 
 its periphery, as in Fig. 17. The vacuolated epithelium has a 
 definite significance in the normal processes of the breast, and 
 there is hardly any doubt that it has a corresponding significance 
 in the pathological. Vacuolation of the epithelium is found in 
 the first of the above four cases from the human subject ; the 
 epithelium is accumulated in several layers on the walls of the 
 acini (Fig. 29), and their lumen, where it is not obliterated, is 
 occupied by a number of cells vacuolated to such an extent as to 
 be little more than hollow vesicles. They are the cells of the 
 upper stratum that have come nearest, as it were, to the proper 
 functional activity, and have been thrown into the cavity of the 
 acinus. The well-known mother-cells of cancer exemplify the 
 same degree of functional change ; sometimes they contain one or 
 more young cells in their vacuoles, but they often occur in the 
 signet-ring form. It has been pointed out by Yirchow that such 
 transformations (degenerations) of cells in cancerous tumours, lead 
 to a partial healing of the disease \ It is only another mode of 
 expressing the same fact that the more or less complete trans- 
 formation of some of its epithelium is the nearest approach to the 
 healthy function that the gland attains to under its diseased cir- 
 cumstances. Atrophic cancer of the breast is a return of the 
 diseased breast to the paths of health. It may be described, 
 also, as the intensifying of a functional excitation that is morbidly 
 weak, or as the raising of the functional force from a dangerous 
 degree of feebleness to a safe degree of strength. 
 
 The above-mentioned cases of mammary tumours in women 
 occurred after the climacteric period. I shall now refer briefly to 
 four cases belonging to an earlier period of life. One of them 
 occurred in a widow aged 36. The disease was at first circum- 
 scribed and situated in an outlying part of the gland ; but from 
 injudicious treatment, the tumour had grown rapidly and pro- 
 truded throucjh the skin as a fungus. The whole breast was re- 
 moved at the operation, and the unaffected part of it was found 
 to have very much the same structure that has been described for 
 
 1 ArcMv, Vol. I. p. 138. 
 
Cn. VII.] OF THE BREAST. 171 
 
 the resting state of the mamma in the bitch. One cluster of 
 greatly enlarged acini occurred in the neighbourhood of the 
 tumour, which showed a distinct approach to the structure of the 
 actual tumour. That structure was of the usual intra-acinous 
 kind ; the cells were almost entirely nuclear, and of a large size, 
 and numerous multi-nuclear cells occurred among them. Besides 
 the accumulations of crude epithelium within the acini, there 
 Avere extensive columns of the same cells in the fibrillar tissue 
 outside. 
 
 Another case occurred in a married woman aged 39. Parts of 
 the gland within the tumour area were found little affected ; 
 they consisted of lobuli in an extreme state of involution, sepa- 
 rated from each other by wide tracts of fibrillar tissue. The 
 lobuli were made up of a number of lateral offshoots or infundi- 
 bula from a central duct ; the cells were of the nuclear kind 
 always found in the resting state of the gland. The growth itself 
 was almost entirely intra-acinous. The acini were not very greatly 
 enlarged, and the cells within them were more of the kind that 
 are found in myxomatous tumours than in medullary cancers; 
 that is to say, many of them were staff-shaped or crescentic 
 elements lying among a homogenous or fluid substance. 
 
 A third case, of Avhich the clinical history is wanting, pre- 
 sumably belonged to the same pre-climacteric period of life. The 
 tumour disease extended uniformly over the whole or the greater 
 part of the gland, and the normal lobular subdivisions could be 
 traced. All the acini were of the same size, and not much 
 larger than the acini of the resting state. They were uniformly 
 filled with groups of highly granular nuclear cells, which in some 
 cases were so closely cemented together by intercellular substance 
 as to resemble single multi-nuclear cells, while in other cases there 
 was a small cavity in the centre of the group. 
 
 In those three cases, the tumours were of the soft kind ; the 
 neighbouring glandular structure, so far as it was examined, had 
 the usual appearance of the periodical resting state ; and in two at 
 least of the cases the patients had not reached the climacteric 
 period. I have one more case to mention, which is intermediate 
 between the first group of post-climacteric tumours and the group 
 just described. It occurred in a widow aged 40, and it was a 
 distinctly hard cancer. The acinous structure could be traced 
 
172 PATHOLOGICAL PROCESSES [Pt. II. 
 
 clearly in the tumour, in the form of alveolar spaces. The cells 
 lay loosely within the spaces and appeared to be suspended in a 
 clear fluid. The forms of the cellular elements resembled those 
 already described for the myxomatous transformations in the 
 mamma of the bitch, corresponding to the mucus-producing stage 
 on the physiological scale. Frequently a large nucleus lay as it 
 were free within a cell with a somewhat thickened wall, and the 
 fragments of these cell-walls or cell-membranes were sometimes 
 found separately in the form of crescent-shaped bodiesS 
 
 The study of these mammary tumours of women would not, 
 of itself, have suggested the principles of tumour-formation in the 
 breast, or the illustrations of morbid physiological action, which 
 I have endeavoured to derive from the examination of the tumour 
 disease in the bitch. But if the appearances in the human 
 cancers be interpreted according to the physiology of the breast, 
 normal and morbid, as derived from the latter source, it will pro- 
 bably be admitted that there is nothing in them that involves any 
 additional or different principle of general pathology; It is the 
 climacteric effacement of the breast that gives a peculiar character 
 to the disease in women ; and there are well-marked structural 
 differences in the tumours according as they appear before or 
 after that period. Those that develope after the climacteric years 
 are perhaps the most common, as they are certainl}^ the most 
 intractable ; they have figured most prominently among the 
 tumours of the breast, and they have been the real source of 
 ambiguity in the pathology of the organ. That ambiguity depends 
 upon the circumstance that they occur in an organ which is 
 gradually losing or has lost its characteristic structure. Where 
 the normal itself is vanishing, the departures from the normal are 
 elusive. The problem becomes all the more complicated when we 
 consider that the effacement of the mammary structure and func- 
 tion, occurring as it does in an organism that is still vigorous in 
 all resj)ects but the sexual, is itself a kind of pathological in- 
 cident, or at least an event that might very easily assume a patho- 
 logical character. 
 
 1 "In Carcinoma alveolare the walla of those old cells that contain the genera- 
 tions of young cells appear at last to break up and to form distinct fibres, which re- 
 main unconnected with one another." — Johannes Miiller, Ueber denfeineren Bau 
 dcr Icranklmftcn Gesclnciilste, p. 5. 
 
Ch. VII.] OF THE BREAST. 173 
 
 It will now be convenient to sum up the more general conclu- 
 sions of the pathological inquiry. From the study of the cases in 
 the bitch, the inference seems justified that the earliest and most 
 general disturbance in the mamma is a kind of spurious evolution 
 from the resting state. The fringes of glandular tissue round the 
 tumours, and the more distant parts when they could be got for 
 examination, were found either in the resting state, or more or less 
 advanced in the process of unfolding. The departures from the 
 resting state were either slight in degree, or, if more extreme, they 
 were limited to small areas. Thus, the production of large granular 
 pigmented cells, indicating the earliest departure from rest, was 
 very generally observed in the tissue near the tumours ; in a 
 few cases the marsfin of grland showed a mixture of all conditions 
 of the acini, from the state of extreme involution to that of full 
 expansion. For example, the two varieties of polyhedric epithe- 
 lium in Fig. 2 are taken from the tissue near a tumour, most of 
 which was in the upfolded state. Out of the whole collection of cases 
 in the bitch, there is only one in which the neighbouring glandular 
 tissue did not show either the resting condition or limited departures 
 from it. The tumour was a large soft growth having the intra- 
 acinous structure of medullary cancer; the fringes of gland that 
 were removed with it had the parenchymatous appearance of the 
 fully expanded organ, and the minute structure of the acini was 
 uniformly that of the most advanced stage of the periodical evolu- 
 tion. If the animal was pregnant, the condition of the mamma 
 would be at once explained ; if pregnancy could not be assumed, 
 the cause of the very regular and extreme state of unfolding is not 
 at all obvious from the pathological point of view. The morbid 
 influence that causes a tumour to form in the breast probably 
 affects the organ elsewhere than at the point of tumour-formation; 
 but the general disturbance of the gland from the resting state is 
 not as a rule carried to any considerable degree. 
 
 A circumscribed tumour arises at a particular part of the gland 
 where the spurious excitation has advanced beyond the earlier 
 stages of evolution or unfolding. In that particular region, pro- 
 bably a territory defined by the blood-vessels, the functional force 
 has acted at a uniform imperfect level for a length of time ; the in- 
 evitable cellular waste of the crude secretion has accumulated within 
 the acini or around them, and the foundation of a tumour has been 
 
174 PATHOLOGICAL PROCESSES [Pt. II. 
 
 laid. The distinguishing feature of the less malignant forms of 
 tumour is that the spurious functional activity comes nearer in 
 the degree of its intensity to that of the perfect secretory force ; 
 the transformation of the epithelium is a more real transformation, 
 and the cellular waste is reduced, in part at least, to the class of 
 fibre-like or crescentic elements that characterise the myxomatous 
 and cartilaginous issues of the tumour process. Some of the 
 atrophic or regressive cancers of the breast in women may also be 
 said to depend on a more real vacuolation of the glandular cells. 
 
 The circumstance that the unknown diseased excitation most 
 common]}'' befalls the gland when it is in the state of rest, is of the 
 first importance in accounting for the formation of a tumour. 
 Whether the disturbance be a mechanical injury, or a sympathy 
 with excitement in the ovaries, or of a more general emotional 
 nature, it comes upon the breast in its resting state. The breast 
 can react in no other way than by following the somewhat slow 
 process of its normal evolution ; without the intermediate stages 
 of vmfolding, it cannot reach the perfect degree of its function in 
 which there would be immunity from danger. The intermediate 
 stages are necessarily associated with the formation of crude cel- 
 lular products; it is at one or other of the intermediate stages 
 that the morbid force delays, and the corresponding cellular se- 
 cretion of the gland thereupon assumes the character of a formative 
 or tumour process. With the effects of a disturbance from the 
 resting state, we may contrast the very different effects of a dis- 
 turbance of the perfect function. When the mamma in its state 
 of full expansion and perfect functional activity becomes the sub- 
 ject of an interference, the result is very commonly a diffuse or 
 nodular inflammation and the formation of abscess. A sudden 
 stoppage of the milk soon after the lactation has been established 
 is apt to produce inflammation, and the same result, or a degree of 
 it, sometimes follows the weaning of the child after a long course 
 of suckling\ The disturbing cause, whatever it may be, acts upon 
 the mamma when its function is at the greatest intensity, and the 
 characteristic effect is inflammation and abscess. The precise nature 
 
 1 ''The mammary glands often become immensely distemled at the time of 
 weaning, and by manipulation bard knots or Inmps are discovered. ..The axillary 
 glands may become painful, swollen, and even advance to suppuration." — Birkett, 
 Diseases of the Breast, p. 182. 
 
Ch. Vir.] OF THE BREAST. 175 
 
 of the cellular processes that go on in inflammation is still a matter 
 of controversy. According to one view, the leucocj^tes that are 
 found to crowd the inflamed area are derived in every case from 
 the blood-vessels, being the escaped white corpuscles ; according to 
 another view, the leucocytes of an inflamed part may have other 
 origins as well. If reference be made to the figure on page 26, it 
 will be seen that the fibrillar tissue about the acini, and the 
 lymphatic channels encircling them are crowded with leucocytes or 
 lymjjhoid cells, which are derived from the secreting epithelium. 
 The case was precisely one of those cases where an inflammation 
 of the gland might have been expected ; the ewe gave birth to a 
 dead lamb at the full time, and the udder remained unmilked. 
 The figure shows the state of the glandular tissue three weeks after 
 the parturition ; there were no actual signs of inflammation, but it 
 is easy to imagine that a more excessive or more general pro- 
 duction of the same nuclear cells from the epithelium would 
 have led to the appearances of inflammation and of abscess ; and 
 as there exists no theory of inflammation of the breast based on 
 observation, such a process, amounting to a catarrhal state of the 
 secreting parenchyma, is as probable as any other. The small 
 round cells in the figure are products incidental to a somewhat 
 enfeebled secretory activity ; they are the characteristic kind of 
 waste product that the epithelium tends to form when the 
 functional excitation is only a little short of its highest intensity. 
 It may therefore be said that a disturbance of the glandular 
 function tends more to transitory inflammations or abscesses and 
 less to permanent cell-collections or tumours, in proportion as the 
 secretory excitation is near its highest intensity and the epithelium 
 of the acini near its perfect form. Contrary to the scholastic 
 maxim in morals, the pathological case may be said to be one of 
 corruptio optimi optima^. 
 
 1 A place may be found here for an observation by Sir Astley Cooper {The 
 Anatomy of the Breast, p. 137). " Suckling also diminishes the disposition to 
 malignant diseases of the breast ; for although women -who have had children are 
 still Uable to cancerous and fungoid diseases, yet it is imdoubtedly true, that breasts 
 which have been unemployed in suckling, hi women who have been married but are 
 childless, and in those who have remained single, are more prone to malignant 
 diseases than those of women who have nursed large families; and if it were only to 
 lessen the probabUity of the occurrence of such horrible complaints and causes of 
 dissohition, women ought not to refuse to suckle their offspring." 
 
170 PATHOLOGICAL PROCESSES [Pt. IT. 
 
 The comparatively safe issue that follows the disturbance of 
 the glandular function at or near its highest intensity, affords an 
 analogy for the comparatively safe or curable tumours of the female 
 breast that are apt to occur between puberty and the age of thirty 
 or thereabout. They are classed by Billroth under the name of sar- 
 coma of the breast, and their characteristic is said to be that they 
 not unfrequently tend to undergo spontaneous resolution*. It is also 
 a familiar fact of surgical practice that the same class of tumours 
 of the breast are sometimes dispersed by strapping or by the 
 application of cold, or by other means. On the other hand, by 
 far the greater number of the really intractable tumours of the 
 breast are those that occur at or near the climacteric years. The 
 occurrence of that period in the life of a woman affects the mamma 
 both in a positive and in a negative way. The final stoppage of 
 the menstruation is known to produce certain "sympathetic" 
 movements in the breasts ; they sometimes even enlarge and 
 discharge a milky secretion. In many w^omen the sudden dis- 
 turbances at the climacteric period are probably greater than the 
 organ has experienced for years before. But the real fatality 
 of those climacteric disturbances, and of all subsequent excitations 
 of the organ, is that the breast then suffers an effacement of its 
 secreting mechanism and a withdrawal of its secretory force. If a 
 morbid influence comes upon the breast within the years of 
 sexual maturity, there is always the possibility that the organ 
 may sooner or later throw off the diseased action by a return 
 to its vigorous normal activity. But if the capacity of func- 
 tional excitation be lost, and the possibility of unfolding finally 
 surrendered, then the played-out organ falls an easy prey to 
 whatever disturbance may reach it through its old physiological 
 channels. If the functioual capacity of the breast did not, in 
 the human species, come to an end long before the natural term 
 of life, there would probably be much less risk of intractable 
 tumour-formation ; and it is only in those tumours that form in the 
 breast within the period when the function may still be awakened 
 to its full and healthy vigour, that a resolution of the disease, or 
 a dispersion of the diseased products, may be looked for. 
 
 In these particulars the liability of the breast to intractable 
 tumours is somewhat special to the human subject. The more 
 1 C li i I itry incite rathohxju', p. 700. 
 
Ch. VIL] of the breast. 177 
 
 general liability to tumour-formation arises out of the general 
 physiological characters of the gland. Its secretion is inevitably 
 cellular throughout a considerable part of its functional existence. 
 The cellular products are incidental to that law of endogenous 
 cell-formation according to which the secretion of the gland 
 appears to be formed ; they are not derived from the epithelium 
 by ordinary generation, but they are products of an alien or hete- 
 roplastic type. A disorder of function, where the normal arrange- 
 ments are of such a kind, leads in the most natural way to tumour- 
 formation. The descri23tions, in this and in the former chapter, 
 are selected illustrations of the various degrees of disordered 
 function, which have resulted in various kinds of tumours. If 
 there be any new and heterodox pathological doctrine involved 
 therein, that result is at least the outcome of a scientific method. 
 The attempt to illustrate pathology by physiology is according 
 to the best precedents, and whatever novelty there may be in these 
 pathological chapters is a novelty that comes from the physio- 
 logical side. 
 
 c. c. 12 
 
CHAPTER YIII. 
 
 TUMOUR-INFECTION OF LYMPHATIC GLANDS. 
 
 In the two preceding chapters I have endeavoured to show, by 
 references to cellular physiology, that the common tumours of the 
 breast take origin in certain irregularities of the glandular func- 
 tion; the tumour-disease of the breast has been represented to be 
 essentially a disorder of function. From that genetic point of view, 
 the new growths in the breast can hardly be admitted to be things 
 in themselves, capable of being defined per genus et differentiam, or 
 referable to the classes of a tumour-kingdom, after the manner of 
 natural objects in the animal or vegetable kingdoms. To make 
 use of a philosophical term, it is not from the ontological point of 
 view that tumours of the breast have been discussed. But, as 
 a matter of practical convenience, the classification of tumours can- 
 not be abandoned ; and even in theory we are ultimately driven to 
 regard tumours as things in themselves. However closely we may 
 trace the departures from the normal to which a tumour of 
 the breast owes its origin, there is sooner or later manifested 
 in typical cases of the new growth, a property which is not in 
 any sense a departure from the normal of the organ, but an 
 attribute possessed by or acquired by the tumour as a separate 
 individual. That property is the malignity of the tumour ; and 
 the malignity is best seen in the infectiveness of the tumour, or in 
 its tendency to beget its like in the form of secondary tumours 
 throughout the body. The present chapter, relating to the infec- 
 tiveness of tumours as shown in the infection of lymphatic glands, 
 is therefore complementary to the former chapters, in so far as it 
 
Cii.VIlL] TUMOUR-INFECTION OF LYMPHATIC GLANDS. 171) 
 
 recognises the tumours to be entities or individuals that have the 
 remarkable property of begetting their like. 
 
 The investigation on infected lymphatic glands is also related 
 to the rest of the work, in that the process of their infection is as 
 far as possible explained or analysed on the basis of the physio- 
 logical chapter treating of the lymphatic glands of the breast. 
 When the lymphatic glands are infected with the tumour-disease 
 that is primary in the breast, one naturally refers for an explana- 
 tion to the kind of relationship that subsists in health between the 
 secreting organ and its lymphatic appendages. It is not for a 
 moment to be expected that so vast a problem as that of infec- 
 tion will find so simple a solution, but the intelligible relationship 
 between the breast and its lymphatic glands offers a convenient 
 means of approaching it. I shall therefore begin with the case of 
 lymphatic glands infected from tumour of the breast, and after- 
 wards proceed to consider the question generally, taking the facts 
 from cases both of epithelial and connective-tissue tumours. 
 
 The infected lymphatic glands of which preparations have been 
 made, were derived from upwards of twenty different cases. Seve- 
 ral of them were mammary cases, both in the human subject and 
 in the dog and cat. In three cases the primary tumours occurred 
 in the stomach, and in one case the primary seat was the great 
 intestine. The infection of lymphatic glands in cases of sarcoma 
 is sometimes doubted, but there are four cases in this series 
 belonging to that class. One of these, of which I have given 
 a detailed account elsewhere S was a tumour growing from the 
 periosteum by the side of the fifth metatarsal bone ; a large 
 packet of lymphatic glands in the groin were affected, as well 
 as many retroperitoneal glands. Another was a case of tumour 
 of the upper jaw, containing many large multinuclear cells ; in 
 that case also the lymphatic glands within the abdomen were 
 everywhere enlarged and changed into tumour-tissue. The other 
 sarcomatous cases were a scapular tumour (in the dog), and 
 a periosteal tumour of the tibia recurrent in the pelvis. The 
 collection included also a number of miscellaneous cases, of 
 which the most remarkable and the most instructive was a 
 case (in the dog) of liver-tumour that appeared to be pri- 
 mary, with secondary infection of portal and mesenteric glands. 
 1 Journal of Anatomy and Physiology for April, 1878. 
 
 JO 2 
 
ISO TUMOUR-INFECTION [Pt. II. 
 
 Many of the cases had secondary tumours of the liver or lung, or 
 of both organs, and in one case (sarcoma of metatarsus) both ova- 
 ries were infected with the tumour-disease. I have made use of 
 the whole of this material in my investigation, but I shall con- 
 fine myself, in this digest, to the description of a few typical 
 cases. 
 
 The one generalisation that stands out conspicuously from the 
 examination of the whole collection of cases is that the secondary 
 tumours (of lymphatic glands and other organs) correspond in 
 structure to the primary, even in the most minute particulars. 
 There are no doubt some cases in which the identity is not per- 
 fectly obvious; but in the great majority of cases the resemblance 
 or mimicry is carried to so great an extent, that it is not possible 
 to mistake the significance of it. This well-established fact of a 
 minute resemblance between the primary tumour and its secondary 
 or infective products, affords at the outset an invaluable clue to 
 the nature of the infective process as regards tumours ; and the 
 tumour-infective process, marked as it is by definite structural fea- 
 tures, may be justly regarded as an index of infection in general. 
 By establishing a resemblance or family likeness between a pri- 
 mary tumour and its secondary products, we are led to conclude 
 that the iufectiveness of primary tumours is a property that they 
 develope of themselves ; no agency from without can be thought 
 of, which might call into existence an infection carrying with it so 
 particular — one may say, so autochthonous — points of structure. 
 
 In illustrating further the infectiveness of a tumour, it will be 
 useful to revert for a moment to the physiological interpretation 
 that has been applied, in the previous chapters, to tumours of the 
 breast. The primary tumour is the mammary gland, or a portion 
 of the gland, more or less modified in its structure through aber- 
 rations of its function. The functional disturbances are never 
 exactly the same in any two cases, and the respective modifications 
 of structure, or, in other words, the structures of the respective 
 tumours, are never exactly the same. Although there are neces- 
 sarily more points of likeness than of unlikeness in two cases of 
 tumours of the same organ, arising at corresponding periods of 
 life, or under corresponding circumstances generally, yet each 
 tumour has its distinct individuality. There are the same differ- 
 ences amonw individual tumours as there are among individual 
 
Ch. VIII.] OF LYMPUATIC GLAXDS. 181 
 
 persons. Now, it is the individuality of the tumour that is re- 
 produced in the secondary tumour. That fact is so remarkable 
 that the attention is at once arrested by it, and it is no less 
 striking as a phenomenon than significant as a point of argument. 
 If the secondary tumour reflects the individuality of the primary, 
 it is obvious that the infective influence emanates from the primary 
 tumour, and that it is a property of the latter developed within 
 it, and independent of all extraneous influences. Whatever the 
 nature of the infection may be, it appears certain that it is derived 
 in the most intimate way from the primarily diseased part. In 
 other words, the infectiveness of a tumour is a property of the 
 cellular elements of the tumour and of their juices. 
 
 The subject of lymphatic-gland infection has not hitherto 
 engaged a great deal of attention, but there have been a few 
 important contributions to our knowledge of it within the last 
 twenty years. In particular, Billroth has made investigations of 
 the lymphatic glands in a variety of infective processes, including, 
 among others, those of acute inflammation, scrofulosis, t}^hoid 
 fever, and cancerS The infected mesenteric glands in typhoid 
 fever were studied by Virchow, and compared with the lymphatic 
 glands in other infective processes^ The infection of axillary 
 lymphatic glands in a case of cancer of the breast has been 
 described by W. Muller^. The general question of lymphatic 
 gland-infection has also been treated of by Hansen*, with special 
 reference to leprosy and scrofula. A more than usually elaborate 
 account of the infection of bronchial glands in a case of cancer of 
 the oesophagus has been published by Gussenbauer^ It will not 
 be necessary, in the sequel, to refer otherwise than in general 
 terms to the conclusions of these writers. 
 
 In proceeding to the details of this investigation, it will be 
 convenient to take first the cases of mammary tumours in the 
 bitch, where the lymphatic glands, especially those beneath the 
 inguinal portion of the mamma, were affected with the disease. 
 
 ^ Beitrage zur pathologischen Hi^tologie, Berlin, 1858; and in Virclioio's 
 Archiv, Vol. 21. 
 
 ' WUrzburger Verlmndl. i. 86. 
 
 =* Uenle und Pfeufefs Zeitschrift, 1863. 
 
 * In a memoir published at Christiania, 1871. 
 
 ' Archiv far Klinisclie CJiiritrgie, 1872. 
 
182 TUMOUR-INFECTIOX [Pt. II. 
 
 In the third chapter I described at length the nature of the 
 physiological relation between the secreting gland and its lym- 
 phatic glands. The lymphatic gland was a kind of depuratory 
 apparatus, receiving the cellular waste of the secreting gland, and 
 reducing it to the uniform condition of lymphoid cells. The most 
 noticeable and unmistakeable form of waste cells were large 
 nucleated granular cells, containing a yellow or brown pigment. 
 Those cells, being waste products of the imperfect secretion, were 
 conveyed to the neighbouring lymphatic glands in large numbers, 
 and in many preparations they could be seen occupying the lymph- 
 sinuses. The pigmentation of the cells enabled them to be traced 
 easily from their origin within the acini to their final disposal and 
 utilisation in the lymphatic gland. Such, then, being the well- 
 marked relation between the secreting gland and its lymphatic 
 appendage in health, the question naturally arises whether the 
 "infection" of the lymphatic gland from a diseased mamma is in 
 any way dependent on a modification of the health relationship. 
 If that should prove to be so, the infection of the lymphatic glaud 
 would be placed on a different footing from the infection of the 
 liver, as described by me in another paper. The liver substance 
 is nothing more than a convenient soil for the infective process to 
 plant itself upon ; whereas the lymphatic glands near the mamma 
 are not merely dense masses of parenchyma liable to be attacked 
 by the infection, but they are also physiological adjuncts of the 
 primarily diseased organs, and part and parcel of the secretory 
 apparatus. It will therefore be a first subject of inquiry, whether 
 the infection of the mammary lymphatic glands is in any way 
 incidental to the collecting and purifying part that they play 
 towards the waste products of the secreting gland. 
 
 One case may be taken as an illustration of the problem here 
 presented. The tumour occuri'ed in a bitch at the inguinal end of 
 the mamma on one side. It consisted mainly of a thickening of a 
 small part of the mamma and of one or two circumscribed small 
 nodules. In removing the affected part of the mamma, it was 
 found that a cluster of lymphatic glands, lying immediately be- 
 neath the mamma and embedded in fat, were both enlarged and 
 otherwise changed; and they were removed adhering to the 
 tumour along with some of their surrounding fat. Two of them 
 were about half-an-inch lonor and somewhat less in their shorter 
 
Cii. VIIL] OF LYMPHATIC GLANDS. 1S3 
 
 axis. On cutting them open there were differences at once ap- 
 parent in tlieir substance, certain tracts being deeply pigmented 
 and denser, while other parts were free from pigment and 
 more elastic. The microscopic examination showed in various 
 places a definite new formation. The new structure generally 
 existed within the limits of a follicle. There were two varie- 
 ties of it. One of them was a structure composed of large 
 rounded spaces lined with a kind of cubical epithelium with large 
 nuclei. This structure closely resembled the appearances in certain 
 parts of the primary new growth. The new structure in another 
 of the lymphatic glands was different. It consisted of long columns 
 of epithelial-like cells, extending into the central clear space of 
 the follicles with their extremities free, or forming a network of 
 trabecular crossing the area of the follicle. That kind of structure 
 is quite definite and is often found in mammary tumours of the 
 bitch; in this case the primary tumour, or the diseased portion of 
 mamma, did not show precisely those appearances at any part of it 
 that was examined, but it showed a papillary kind of structure 
 that had doubtless grown out of an earlier form such as that found 
 in the lymphatic gland. 
 
 Leaving as it stands the question of resemblance between the 
 primary structure and that in the lymphatic glands, we may turn 
 to consider how the new structure in the lymphatic glands was 
 acquired. The glands, it has been said, were obviously pigmented 
 more at some parts than at others. In the normal condition, the 
 same lymphatic glands are often found to have their lymph-sinuses 
 occupied by large granular pigmented cells, and the follicles are 
 sometimes found to be strewn with grains of free pigment. In 
 this pathological case, the most noticeable point of difference is 
 that the large round pigmented cells have overflowed from the 
 lymph-sinus into the substance of the follicle. But there are 
 certain other changes in the diseased gland which may or may not 
 be antecedent to, or the cause of, the overflow of mammary cells 
 from the lymph-sinuses into the follicles. The lymph-sinuses 
 themselves are often seen to be more obstructed by their contents 
 than normally. The fibres stretching across them and forming 
 their filtering or depuratory apparatus are broader and thicker, 
 and the pigmented cells that ordinarily pass through them, and 
 become stripped of their pigment and cell-substance in so doing, 
 
18 i TVMOUR-INFECTION [Pt. II. 
 
 appear to have taken up as if permanent positions ; the lymph- 
 passage is in fact clogged up. There is, however, a more general 
 change in the parenchyma of the gland, of which the change in 
 the lymph- sinuses is perhaps only a part. Throughout the whole 
 of the gland the lymphoid cells have become swollen. In ordinary 
 circivmstances, and under ordinary methods of preparation, the 
 lymphoid cells appear as naked nuclei. In the beginning of the 
 infective cancerous process, they are all found to have a narrow but 
 distinct investment of protojDlasm; the nucleus is now surrounded 
 by a margin of cell-substance, and the closely-packed cells of the 
 follicles have often a rectangular or polyhedric shape, resembling 
 epithelium. The swelling of the parenchyma is quite general 
 throughout the gland, and it is difficult to say by what agency, 
 issuing from the seat of primary disease, it may have been brought 
 about. The swelling of the parenchyma of the secondarily infected 
 organ appears to be the first step in the process of cancerous 
 infection. 
 
 The swelling of the lymphatic parenchyma is not peculiar to 
 the infective process of malignant tumours. It occurs, generally 
 speaking, in every infective process, including acute and subacute 
 inflammation, scrofula, tuberculosis, syphilis, typhoid fever, &c. 
 The resemblance in their initial stage between certain of these 
 processes was made out by Virchow, and between others by Bill- 
 roth; so that it may be said to amount to a generalisation. In 
 the cases of infected mammary lymphatic glands, it is difficult to 
 decide whether the swelling of the lymphoid cells of the paren- 
 chyma preceded or followed the overflow of cells from the primarily 
 diseased part into the follicle, and it remains therefore doubtful 
 how far these cells, by their contact or otherwise, were the agents 
 of infection. But whatever the cells brought from the diseased 
 mamma may have had to do with the initial swelling of the 
 parenchyma, which is common to this particular infective process 
 and to every other infective process, there is little doubt that they 
 sometimes determine the permanent structural changes in the 
 cancerously-affected lymi^hatic gland. 
 
 To return to the case of lymphatic gland beneath the diseased 
 mamma of the bitch. The new growth in the lymphatic gland 
 consisted, in some places, of papilliform processes or trabeculoe of 
 epithelial-like cells, resembling a structure of common occurrence 
 
Ch. YIIL] of lymphatic G lands. 185 
 
 in primary mammary tumours. lu tlie primary disease, tliat struc- 
 ture is dependent on outgrowths of the epithelium lining the acini; 
 several varieties of this primary process in the mamma were de- 
 scribed and figured in the sixth chapter. In the secondarily 
 affected lymphatic gland, the same papilliform processes and trahe- 
 culjE occur, and they appear to arise from the wall of acinus-like 
 cavities, just as they did in the acinous gland. How then does the 
 lymphatic gland assume this wonderful imitation of the pathologi- 
 cal process in the secreting gland? 
 
 In the first place, it is the dense parenchyma of the follicles and 
 cylinders that is the seat of the changes; the lymph-sinuses may 
 be disregarded for the present. The lymphoid cells of the follicles, 
 it has already been pointed out, have acquired an investment of 
 protoplasm or a fringe of cell-substance round the nucleus. As in 
 the first stage of all the infective processes, the lymphoid cells have 
 acquired something of an epithelial character. In other infec- 
 tive processes the swelling of the parenchyma is only a passing 
 phase ; the swollen cells undergo rapid changes either in the way 
 of suppuration or of necrosis, and the process, if it runs its course, 
 ends in cicatrization. The tumour-process is eventually dis- 
 tinguished from all of these; although in some of the cases of 
 tumours, to be afterwards mentioned, there is a resemblance with 
 other infective process beyond the point of parenchymatous swel- 
 ling. In the particular case, however, that has been selected for 
 the general description, the lymphoid cells permanently retain their 
 protoplasmic investment, and become endowed from that time for- 
 ward with the persistent vitality and power of increase which dis- 
 tinguishes tumour-cells. 
 
 Up to this point we have found that the cells of the whole fol- 
 licle become uniformly changed into epithelial-like cells. It is out 
 of that matrix-tissue that a particular structure is carved, according 
 to the pattern of the primary tumour. The compact follicle be- 
 comes changed into an acinus-like cavity, with papillary processes 
 rising from its margin (or its floor), and projecting free into its 
 space, or with trabecular bands crossing it in various direc- 
 tions. The carving out of the tumour-pattern from the compact 
 cells of the follicle is effected by means of the large granular pig- 
 mented cells. These, it has been mentioned, overflow from the 
 lymph-sinuses into the follicles. They are frequently seen lying 
 
18G TUMOUR-INFECTION [Pt. II. 
 
 among the cells of the follicles, and in favourable specimens their 
 agency in transforming the follicle into a portion of new growth, 
 after the primary pattern, can be inferred in a general way. Many 
 of them become distended, a cavity forming in their substance ; 
 they are thus transformed into large vesicles, which occupy cor- 
 responding spaces among the follicular cells. The vesicles then 
 break down^ and the follicular cells remain as epithelial-like pro- 
 cesses or trabeculse, wuth spaces between them, and looking as if 
 they had grown up from the rows of cells lining the wall of a 
 cavity. In the primary' tumour the round spaces, with numerous 
 knob-like or papillary projections, are actual acini of the gland, 
 the lining epithelium of which has taken on the common form of 
 papillary overgrowth. Exactly the same appearance is produced 
 as a secondary result in the adjacent lymphatic gland ; and the 
 striking resemblance, or rather the exact copy, is effected by carv- 
 ing the pattern out of the compact mass of cells of the follicle. 
 The carving-out process is accomplished by means of a particular 
 class of cells, which are themselves destroyed in the process. Now 
 the large pigmented cells, through whose agency the pattern of the 
 primary tumour is imprinted on the secondarily affected lymphatic 
 gland, are derived from the primary tumour, or, in other words, 
 are products of the primarily diseased organ. 
 
 That is, however, only one of the ways in which the lymphatic 
 gland is transformed into the likeness of the primary tumour. 
 Another mode, which is an interesting modification of the first, is 
 as follows. It has been observed especially well in the affected 
 lymphatic glands of a case of mammary tumour of the human sub- 
 ject. The primary tumour in this case is figured on page 143 and 
 again on page 144. It is essentially of the same trabecular structure 
 as the mammary tumour of the bitch just described; the enlarged 
 acini are traversed by anastomosing columns of cells. The structure 
 of the axillary lymphatic gland is of exactly the same kind; there 
 is observed in the primary tumour at certain places a quantity of 
 brown pigment-granules, and collections of red blood-corpuscles, 
 and even in these small and, as it were, accidental circumstances, 
 the lymphatic gland resembles the growth in the mamma. The 
 development of the new growth can be traced through various 
 stages. The first stage is that of general sw^elling of the paren- 
 chyma, depending not so much in this case on a distinct fringe of 
 
Ch. VIII.] OF LYMPHATIC GLANDS. 187 
 
 protoplasm rouud the previously naked nuclear lymphoid cell, as 
 upon a uniform enlargement of the lymphoid cell in its nuclear 
 state, the cell becoming at the same time more granular. The 
 next stage resembles that found in other infective processes. 
 The swollen parenchyma appears to undergo a general decay or 
 necrosis. The swollen cells assume an opaque appearance, and 
 fail to colour with the logwood staining fluid. In the other in- 
 fective processes that stage goes on to the general breaking down 
 of the parenchyma, but in the tumour process the sequel is diflferent. 
 Along certain regular lines throughout the follicle the opaque 
 and, as it were, necrotic cells show a small and brightly-coloured 
 nucleus or nucleolus in their centre. These nuclei are in many 
 cases quite minute points, and such is no doubt their first appear- 
 ance as young cells in the substance of the old follicular cells. 
 They appear, as has been said, along certain regular lines; and 
 growing up from the old cells along the lines that are as if laid 
 down or marked out for them, they result in a new growth of a 
 definite pattern, the pattern being the structure of the primary 
 tumour. There is in this case, as in the former case, a carving out 
 of the new growth from the compact cells of the swollen follicle, 
 but the carving out is here afifected by a residual process. The old 
 cells of the follicle decay except along the lines of the new pattern ; 
 and along these lines young cells spring from the substance of the 
 old follicular cells, and in course of time gTOw up into the form and 
 arrangement of the cells in the primary structure. 
 
 A third example of the mode of transformation of the lymphatic 
 gland-substance is found in the epigastric glands in a case of cancer 
 of the stomach. In the primary tumour the epithelial-like cells 
 were collected in alveolar groups in the submucous and muscular 
 coats. The cells lay in the alveolar spaces, packed together with- 
 out any veiy definite arrangement. The appearance of the primary 
 structure is simulated in the lymphatic gland in the following way. 
 There is first of all the swelling of the parenchyma; the swollen 
 lymphoid cells then become opaque; and ultimately there arise 
 from them, not along particular lines, but over continuous areas, 
 small young cells, which form the alveolar groups corresponding to 
 those of the primary tumour. But the special interest of this case 
 is in showing the exact mode of origin of the young cells from 
 the opaque and decaying follicular cells. The mode of origin 
 
188 TUMOUR-INFECTION [Pr. II. 
 
 is liere clearly seen to be that of endogenous cell-formatiun. The 
 young cell lies in a cavity of the parent cell ; the smallest of them 
 are minute points like a small nucleus, but even at that stage they 
 seem to lie free in a cavity or vacuole. All the lymphoid cells of 
 the follicle appear to undergo this transformation, and the pale out- 
 lines of the old cells can always be seen lying among the en- 
 dogenous brood, even when the latter have reached a considerable 
 size and become independent. I have already described this 
 variety of endogenous cell-formation in my former paper, treating 
 of the development of secondary tumours in the liver. In the 
 present case of cancer of the stomach, there were affected lymphatic 
 glands of various sizes. It is in the smaller glands that the 
 histogenesis of the new growth can be traced; in a large pig- 
 mented gland, the size of a pigeon's Q^g, the tumour-structure was 
 perfect and finished, and the cells had everywhere their epithelial 
 form and arrangement. 
 
 The three cases hitherto mentioned are cases where the sem- 
 blance of the primary tumour was produced in the lymphatic gland 
 by a process of adaptation or carving out of the lymphoid paren- 
 chyma. I shall add a fourth case to illustrate the extreme form 
 of this mode of infection. It is a case of a large sarcomatous 
 tumour in the dog, growing from the ventral surface of the 
 scapula. The brachial plexus of nerves was carried inwards by 
 it, and spread out over its axillary surface, where the nerves, 
 widely separated from each other, appeared to bind it down 
 like the cording of a bale. The lymphatic glands were embedded 
 in the substance of the tumour at its axillary periphery. They 
 could be shelled out from the tumour proper, and they were also 
 distinguishable by their colour and density. There were at least 
 half-a-dozen of these embedded lymphatic glands, and they were 
 upwards of half-an-inch in their longest axis. On examining them, 
 they were found to be uniformly affected, and to be changed 
 throughout into the likeness of the sarcoma which had enclosed 
 them. In some, the process of transformation could be traced ; in 
 others the process was complete, the individual cells having lost 
 everywhere their lymphoid character, and assumed the size and 
 form of the tumour-cells. So far as the preparations went, the 
 process seemed to have befallen the lymphatic glands uniformly 
 throughout their substance. The lymphatic glands were, for the 
 
Ch. VIIT.] of lymphatic glands. 189 
 
 infective process, notliing else than compact cellular masses pre- 
 senting at all points an equally favourable soil for the infection. 
 The case may be taken as an example of those in which the 
 physiological arrangements of the gland, the relation of the lymph- 
 sinuses to the follicles, and the like, go for nothing. The lymphatic 
 gland falls a prey to the infective process as if it had been a crude 
 mass of cells, and as if its complex structure and its perfect 
 mechanism were non-existent. The infection of inguinal and 
 retroperitoneal lymphatic glands in a case of sarcoma of the 
 metatarsus is of the same sweeping and direct kind. 
 
 In such cases it appears hopeless to follow the path of the in- 
 fection, or to inquire whether it came to the lymphatic gland in the 
 form of a fluid or in the form of cells, or to attempt to analyse the 
 nature of its action on the lymphatic parenchyma. But if the mode 
 of infection in the last of the four cases, and in other cases, chiefly 
 sarcomatous, that might have been given, seems to bafile all 
 analysis and to be referable only to an inscrutable agency, it 
 borders on modes of infection in other cases that are less and less 
 mysterious. ' The four cases given as examples all show that the 
 infection acted upon the predominant cells of the gland, or upon 
 the lymphatic parenchyma ; and in the first of them the coopera- 
 tion, in the infective process, of cells derived from the primarily 
 diseased part, or the primary tumour, was undoubted. It is this co- 
 operating or determining agency that has to be further inquired into. 
 
 In the first of the four cases, the cells derived from the primarily 
 diseased part were large granular pigmented cells. Those cells are 
 normal products of the mammary secretion at its periodically en- 
 feebled periods, and in the tumour processes of the bitch they are 
 produced in unusual quantity at unusual times. They some- 
 times contribute largely to the bulk of tumours, being collected in 
 heaps outside the secreting acini; and when they are stripped of 
 their pigmented cell- substance, their nuclei subdivide and multipl}^ 
 in the usual way. In the healthy condition of the mammary 
 function, they are conveyed to the neighbouring lymphatic glands, 
 in whose lymph-sinuses they are reduced to a nuclear or lymphoid 
 condition, and are so utilised as lymph-corpuscles. Now. in the 
 pathological states of the mammary function, as they may be 
 studied in the l)itch, the lymphatic glands may continue to 
 perform this task for the waste products of the secreting gland 
 
190 TUMOUR-INFECTIOX [Pt. II. 
 
 for a considerable time after the pathological process in the 
 mamma has been firmly rooted. In the present collection of 
 niammarv^ tumour-cases there are several in which the lymphatic 
 glands of the mamma, both at the inguinal and axillary ends 
 of the gland, show the normal relations of all their parts ; the 
 follicles contain the ordinary lymphoid cells, and the sinuses are 
 filled with the large yellow mammary cells. But it appears that 
 a time comes when the depuratory apparatus of the lymphatic 
 gland breaks down. 
 
 The first evidence of failure is stagnation of the large gTanular 
 cells in the lymph-sinuses, and that is followed by an overflow of 
 them into the substance of the follicles. This overflow may be 
 observed in many cases, and there is no doubt of its close connex- 
 ion with the process of infection. Side by side with the overflow, 
 there is observed a change in the constant elements of the lym- 
 phatic gland. That change is the parenchymatous swelling before 
 mentioned, and it is difficult to say whether the parenchymatous 
 swelling of the constant elements of the gland is the cause or 
 the effect of the overflow of waste cells that are in process of 
 passing through it. The swelling aff'ects not only the lymjihoid 
 cells of the follicular tissue, but much more the "fibres" of the 
 lymph-sinus. These fibres are protoplasmic bands, and the nuclei 
 that they appear to contain are merely the circulating cells of 
 the lymph-sinus in more or less close apposition to them. At the 
 same time that the lymphoid cells of the follicular tissue be- 
 come individually larger or more invested with cell-substance, the 
 '■fibres" swell up into broad bands, dimiuishing the meshes 
 between them. The lymph-sinus thus appears as if it were 
 clogged, and the depuratory process by which the waste cells are 
 reduced to a nuclear state appears as if arrested. It is then that 
 the cells brought from the mamma overflow into the follicles. The 
 real difference between this pathological process and the normal 
 process is that, in the latter, the waste cells of the secretion are 
 received into the follicle only after they have undergone the 
 depuratory and reducing process of the lymph-sinuses; while in the 
 pathological case they enter the parenchyma of the gland in the 
 form of waste cells, and with all the properties of waste cells cling- 
 ing to them, whatever those properties may be. Having reached 
 the follicles in the condition in which thev left the mamma, they 
 
Cii. YIII.] OF LYMPHATIC GLANDS. 191 
 
 there play a distinct part in producing the semblance of the 
 primary tumour in the lymphatic gland. 
 
 In the first case mentioned, the foreign cells determined the 
 pattern of the new growth, or aided in the carving out of that 
 pattern, while they themselves did not go to make up its bulk. 
 But there is reason to think, from a number of other cases, that 
 the waste cells from the primarily affected part may themselves 
 contribute by their own substance to the production of the patho- 
 logical structure in the lymphatic gland. 
 
 The two modes of infection have been discussed by previous 
 writers, and have been broadly contrasted, and treated as support- 
 ing rival theories. Either the lymphatic gland is transformed 
 throughout its whole substance into the likeness of the primary 
 tumour, or certain fragments from the primary tumour are carried 
 into the lymphatic gland where they find a lodgment and grow up 
 into a tumour-mass, gradually encroaching upon and causing the 
 atrophy or disappearance of the lymphatic tissue proper. The four 
 cases already given as examples are all illustrative of the first 
 theory; but other cases might be given that would appear to 
 support the second. 
 
 The same apj^arent discrepancy or want of agreement occurs 
 throughout the entire literature of this subject, and not only as 
 between different observers, but as between the different observa- 
 tions of the same observer. Thus Professor Billroth, in his earliest 
 papers on this subject, wrote : — "As regards the extension of cancers 
 by the lymph-stream, I cannot entirely share the opinion of those 
 Avho here assume a direct transport of cancer-cells, which are 
 merely conveyed to another j)lace, there to fulfil their destiny. 
 This view is to me somewhat incomprehensible; I might almost 
 say, it is conceived too much in the sense of plant-life. In the 
 widest estimate that I can form of the vital powers of cells, I can- 
 not imagine that they swim in the lymph-stream, like fish in the 
 water, or that, like wandering spores, they require only to fix them- 
 selves in order to grow up into new individuals. It is much more 
 conceivable that the serum (intercellular substance) convej'ed from 
 the cancer has a certain infectiveness for other cells, and that thereby 
 a new diseased process arises in a neighbouring 2Dart\" Profes- 
 sor Billroth returned to the subject in a later investigation, in which 
 1 Bcitrogc zur patlwhgischen Ilistologie, p. 201-2. 
 
192 TUMOUR-INFECTIOX [Pt. II. 
 
 lie bad the advantage of newer observations on tbc lympb-circu- 
 lation tbrougb tbe gland ; and in a case of pigmented cancer of tbe 
 breast he found melanotic cells in the lymph-sinuses of the axillary 
 lymphatic glands, and was disposed to conclude that these were the 
 starting point of the secondary new growth. He especially remarks 
 upon the value of pigmented cells in tracking the infective process. 
 "B}'- making out," he continues, "that the cancer- formation in the 
 lymphatic glands follows the lymph-sinus, that is to say, the normal 
 lymph-stream, we are again brought face to face with the notion 
 of a transporting of cancer-cells. It appears extremely plausible, 
 from this discovery, that the cells from the cancer reach the 
 lymphatic glands by the lymphatic vessels, and that they there fix 
 themselves and establish a new centre {HeercT). But proof is 
 wanting that cells of this kind can be transported, or that they are 
 actually found in the afferent lymphatic vessels of the glands" 
 
 In the present collection of cases are several in which the new 
 growth begins in the lymph-sinuses. Thus, in a lymphatic gland 
 beneath a mammary tumour in the cat, the new structure is found 
 at only a few points in the cortical lymph-sinuses; at these points 
 the fibres of the reticulum appeared to constitute the framework 
 of the new growth. In a lumbar lymphatic gland infected from 
 cancer of the rectum, many cells, that can be traced into their 
 ultimate form of tumour-cells, lie in the lymph-sinuses, and where 
 collections of them occur in the follicular tissue, the collections 
 are always definitely circumscribed. Again, in the case of lym- 
 phatic glands infected from one of the cases of cancer of the 
 stomach, there may be seen in different preparations, various de- 
 grees of encroachment of the tumour-cells from the lymph-sinuses 
 upon the follicles. But many of the lymphoid cells of the follicles 
 retain their lymphoid character to the last, and narrow tracts of 
 them can here and there be seen extending through a gland that 
 is otherwise completely changed into the likeness of the new 
 growth. 
 
 In a large collection of infected lymphatic glands from various 
 kinds of tumours, there is no doubt a considerable want of uni- 
 formity in the process of infection. But it may well be doubted 
 whether the appearances can be placed in two distinct categories, 
 or be taken to establish two distinct processes. The discrepancies 
 
 ^ Virchoiv's Archiv, Vol. xxi. pp. 411-2. 
 
Ch. VIII.] OF LYMPHATIC GLAiWS. 193 
 
 between the accounts of different observers and between the differ- 
 ent accounts of the same observer, depend, without doubt, upon 
 differences really existing. The inference from them, at the 
 utmost, would be that there is more than one mode of infec- 
 tion. But a closer examination seems to show that the process of 
 infection in lymphatic glands is in all cases one and the same. 
 The differences depend upon the greater or less degree of intensity 
 of the infection. Those differences in intensity are often exem- 
 plified in the same case. It is not uncommon to meet with several 
 lymphatic glands affected, and these of various .sizes, say from the 
 size of a pea to that of a pigeon's ogg. The smaller glands contain 
 more of the normal structiu'e than the large, and from that cir- 
 cumstance it might appear as if the larger glands, as found, had 
 been the first attacked, and that in them the new growth had 
 spread farthest. But the most usual difference between the larger 
 and smaller glands appears to be that the larger had from the 
 beginning undergone a more extensive or a more general initial 
 swelling of their parenchyma, whereas the smaller are only affected 
 at limited spots. And that difference may be said to depend on 
 the initial intensity of the infection. 
 
 It is by keeping this circumstance in view that the apparently 
 irreconcileable varieties of the infective process can be shown to be 
 one and the same. There is not one class of cases in which there 
 is a general transformation of the lymphatic tissue into tumour- 
 tissue, and another class of cases in which the tumour-tissue of the 
 lymphatic gland is derived from the multiplication of primary 
 tumour-cells conveyed thither. But there are extreme ca,ses in 
 which the whole parenchyma of the gland appears to be infected 
 at once by some infective agency which cannot be traced; there 
 are less extreme cases where the whole follicular tissue is trans- 
 formed by means of cells overflowing into the parenchyma from 
 the lymph-sinuses; and there are cases of feebler infection where 
 the lymph-sinuses are the starting point, and where the transfor- 
 mation proceeds from them throughout the follicle as if in suc- 
 cessive zones or from successive centres. The last-mentioned cases 
 are those in which the new growth has seemed to depend on a 
 multiplication of the transported primary cells, and a gradual en- 
 croachment upon and supplanting of the normal lymphatic tissue 
 by them. There is no doubt an encroachment; but it is not a 
 c. c. 13 
 
194 TUMOVR-IXFFX'TIOX [Pt. II. 
 
 supplanting and displacing encroach meut, but a transforming en- 
 croachment. 
 
 It will now be necessary to give a digest of the detailed obser- 
 vations on which the above conclusion is based. In the first of the 
 group of four cases, it was stated that the new growth was carved 
 out of the swollen parenchyma or follicular tissue, by the agency of 
 large granular pigmented cells that had overflowed from the lymph- 
 sinuses. More particularly, it was by the vacuolation of these 
 foreign cells that the lymphatic parenchyma was adapted, or trans- 
 formed into the pattern of the new growth. It is to be observed 
 here that there was no vacuolation of the individual cells of the 
 lymphatic parenchyma. The vacuolation or endogenous cell-for- 
 mation process has been shown in my former paper to be the mode 
 of transformation of the liver parenchyma in secondary tumours, 
 and the same vacuolation of the individual cells of the lymphatic 
 parenchyma is now to be described for certain of the lymphatic 
 , glands. In the first case, there was a process of vacuolation, but 
 it took place in the foreign or transported cells; and as the proper 
 cells of the parenchyma were not transformed by that process, the 
 foreign cells overflowing into the parenchyma may be imagined 
 to have undergone the vacuolation process, as it were, instead of 
 the cells of the part. 
 
 The vacuolation process, as an element of the process of tumour- 
 infection, can be nowhere seen more perfectly or more unambigu- 
 ously than in certain cases of infected lymphatic glands. Vacuolated 
 cells are found in lymph-sinuses, while the follicles are still in their 
 lymphoid condition. Again, the large protoplasmic cells in which 
 the vacuolation process takes place are found to have overflowed 
 into the follicles. Sometimes the vacuolating cells form an outer 
 zone to the follicle, at other times they form definite groups in the 
 follicular substance. The products of the vacuolation are the 
 tumour-elements. In some cases a numerous brood of young cells 
 is found within the cavity of the parent-cell. In lymphatic glands 
 affected in cases of cancer of the stomach and cancer of the rectum, 
 the multiplication of nuclei within the cavity of the parent-cell 
 can be observed. It is obviously the same process as that de- 
 scribed as follows by Virchow for scrofulous glands^: "In the 
 lymphatic glands the first stage of the process is a very remark- 
 
 ^ W"t(rzhur<jer J'erlunidhnigoi, Tol. i. p. 85. 
 
Ch. VIII.] OF LYMPHATIC G'LAXDS. 195 
 
 able liyportropliy of the elements by endogenous formation of 
 nuclei, Wi)ilo the cells increase in size five or six times, it is seen 
 that their nuclei multiply in pairs, probably by a division pairwisc, 
 so that as many as twelve pairs of various sizes may be counted in 
 the same cell." It may be mentioned that the largest number of 
 endogenous nuclei in single cells occurred, in the present series of 
 preparations, in a case of secondary tumour in the lung of the cat, 
 after cancer of the breast. The cells lay within the lung alveoli, 
 and appeared to be the swollen cells of the alveolar epithelium. 
 A large endogenous brood is however the exception. The most 
 common case is where the vacuolated cell carries an oblong or 
 crescentic nucleus-like body on one side. It is this body, be- 
 cominsf disengaged, that forms the tumour element. Now, the 
 vacuolation or endogenous process is sufficiently characteristic to 
 be traced. It is first seen in large protoplasmic cells lying in the 
 lymph-sinuses; it is then seen in groups of the same cells at scat- 
 tered points in the midst of a follicle; and lastly, it is seen all over 
 the follicle. In the last-mentioned stage, it is quite obvious that 
 the cells generally undergoing the vacuolation process are none 
 other than the enlarged Ij'mphoid cells themselves. The lymphoid 
 cells swell up, become granular and vacuolate, and thus the trans- 
 formation into tumour-cells becomes general. 
 
 But the question of greatest interest remains. Do all the cells 
 that are seen in the infected lymphatic gland undergoing the 
 vacuolation process, belong to the pre-existing elements of the 
 gland, or are some of them conveyed to the gland from without ? 
 Again, if there are foreign cells among them, how are they 
 distinguishable from, and what is their relation to, those cells of 
 the part that undergo the same process ? It is here that it seems 
 possible to reconcile the discrepancies in the different accounts of 
 Ijanphatic gland-infection. 
 
 As regards the lymphatic glands infected from mammary tu- 
 mours in the bitch, it is possible to speak definitely on the point 
 now at issue, for the large round cells filled with yellow or brown 
 pigment have been shown, in the third chapter, to be normal 
 waste products of the mammary secretion. In health these waste 
 products are conveyed to the lymphatic gland and are there dis- 
 posed of; in disease they are conveyed to the lymphatic gland and 
 play a conspicuous part in the infective process. Their agency in 
 
19G TUMOUR-INFECTION [Pt. II. 
 
 carving out the pattern of the primary tumour from the swollen 
 substance of the lymphatic gland has been already described. 
 Leaving the mammary cases for a time, let us consider cases of the 
 stomach and intestine as illustrating a variety of the process. In 
 the affected lymphatic glands in cases of cancer of the stomach 
 and cancer of the rectum, there are found cells exactly analogous 
 to the large yellow cells in the mammary cases in the bitch. They 
 arc large rounded cells, with the peculiar protoplasm of the glandu- 
 lar epithelium of the digestive tract. In the case of the stomach, 
 there is every reason to think that they are collateral cellular pro- 
 ducts of the secretion, in the same way that the large yellow cells are 
 waste products of the mammary epithelium. It is of such epithelial 
 waste cells, or bye-products of the secretion, that the tumours of 
 secreting structures are composed ; and the cells that are conveyed 
 to the lymphatic gland are not so much tumour-cells, but cells of 
 the primarily diseased part, which, if they had remained in their 
 habitat, would have become tumour-cells. They would have be- 
 come elements of the primary tumour after undergoing a more or 
 less obvious transformation, and they become elements of the 
 secondary tumour by fulfilling the same destiny in their new posi- 
 tion. So far, there is no infection, but a mere change of locality. 
 But their presence in the lymphatic gland leads to a transformation 
 of the cells among which they lie. Whenever the imported cells 
 overflow into the follicles, the cells of the follicle begin to undergo 
 the tumour-transformation. There are two very instructive 
 varieties in this process, which have been adverted to already. In 
 the one, the foreign cells overflow, and are the agents of carving 
 out or determining the exact copy of the primary tumour. This 
 they do by becoming vacuolated and leaving a space in their room, 
 the cells of the lymphatic gland undergoing all the while a direct 
 or non-vacuolar transformation. The foreign cells contribute 
 nothing to the new growth, except its form. The other variety 
 is in strikino- contrast to this. The foreign cells overflow into the 
 follicles, and there, as well as in the lymph-sinuses, they undergo 
 that endogenous process by which they become the elements of the 
 new growth. They induce the same endogenous process in the 
 lymphoid cells near them, and both contribute pa'*'i passu to the 
 buililing up of the substance of the new growth. In this case there 
 is direct vacuulatiun of every single cell ; but in the other case the 
 
Cu. VIII.] OF LYMPHATIC GLANDS. 197 
 
 vacuolation takes place only in the foreign cells, and it may be 
 held, even if it he no laorc than a fancy, that the vacuolation of a 
 few cells stands for the vacuolation of the whole number of cells, 
 so that the transformation of the latter is not round about, but 
 direct. In the one case, also, the initial parenchymatous swelling 
 of the lymphatic follicle took place all over the gland at once ; but 
 in those cases where the individual vacuolation of the follicular 
 cells is observed, the parenchymatous swelling appears to break 
 out in limited zones or islands of the gland, and to be in fact the 
 immediate precursor of the vacuolation process in each cell. 
 
 The simplest mode of infection may therefore be said to be that 
 in which a considerable number of cells are conveyed from the 
 primarily diseased part to the lymphatic gland, where they them- 
 selves undergo a transformation to tumour-cells, and at the same 
 time induce, by contact or by some other unknown influence, a 
 similar transfoi'mation in the proper cells of the part\ The case 
 of gonorrhoeal pus conveyed to the conjunctiva and setting up 
 a purulent formation there, is so far analogous. The analogy 
 appears to hold good to this extent, that as gonorrhoeal pus pro- 
 duces gonorrhoeal pus, so tumour-cells produce tumour-cells; the 
 chief difference being that gonorrhoeal inflammation is morpho- 
 logically a simple process, while a tumour-process implies the for- 
 mation of a more or less complex organoid growth. 
 
 The mode of infection that comes next in complexity and in 
 obscurity is where a few cells overflow into the lymphatic paren- 
 chyma, and determine the transformation of all the cells of the part 
 within a certain radius. There are two instructive points of 
 difi'erence between this case and the last. The few cells that are 
 the agents of infection do not appear to contribute to the substance 
 of the secondary formation, and the cells of the part which alone 
 form the new growth do not, in some cases at least, undergo the 
 characteristic endogenous transformation. But the few infecting 
 cells undergo this transformation, and they may be said to take 
 the transformation upon them, as representing the whole number 
 
 1 With regard to the large quantity of infecting si;bstauce, it is worthy of notice 
 that the seminal material, which may be compared to an infective substance, 's not 
 produced in merely infinitesimal quantities, but generally in great abunc'!auce. 
 See Darwin's Hypothesis of Pangenesis, in hi.; Varial'wn of Animals and I'hnitt:, 
 Vol. II. p. 303. 
 
198 TUMOUR-INFECTIOX [Pt. 11. 
 
 of cells of tlic part. Through the efficacy of the imported cells, the 
 lymphoid cells of the lym])hatic gland assume more or less directly 
 the structure of the primary tumour. Now, it is this direct trans- 
 formation of the secondary -parenchyma into the likeness of the 
 primary tumour, that is characteristic of some of the most malignant 
 or intense tumour-infections, and it may be said to distinguish 
 sarcomatous tumours as a class, where those tumours do have the 
 lymphatic glands affected. In such cases it is difficult to trace any 
 transport of infecting cells. The whole parenchyma appears to 
 transform itself directly into the likeness of tiie parent tumour, as 
 if through some unseen and inscrutable agency. But that agency 
 need not be assumed to differ in kind from the vicarious or 
 substitutive agency of certain cells in other cases, which appears to 
 be associated with the directness of the change into the tumour- 
 tissue. If a few cells originating in, and physiologically belonging 
 to the primarily diseased part, can determine the transformation of 
 a lymphatic gland by being conveyed to it, but without contributing 
 by their own substance to its new structure, then it is conceivable 
 that a similar influence may operate on the lymphatic j^arenchyma 
 from a still greater distance, that is to say, from the original 
 habitat of the cells in the primarily diseased part. Certain it is 
 tliat in tumour-infection we are warranted in assuming no other 
 agency than is contained among the properties of the cell, however 
 obscure those properties may be. The mode of action of the sper- 
 matic elements is a mystery large enough to cover all the varieties 
 of infective influence residing in cells, and to make them seem in- 
 telligible by comparison. But the problem of tumour-infection 
 presents more points of approach than does the mystery of 
 generation. The spermatic influence is, in its effect, one and 
 indivisible, while the infective influence of parent-tumours may 
 vary infinitely in degree. It is from such of these variations as 
 have come under notice in the present investigation that a certain 
 amount of explanation appears to have been reached. In the 
 simplest form, glandular cells of the mamma or the stomach, which 
 would have in time contributed to the primary new growth, are 
 conveyed, by the physiological channels, to the lymphatic gland ; 
 in their changed situation they not only fulfil their destiny and 
 become tumour-cells, but they induce a like transformation in 
 lymphoid cells that they come in contact with. In the next degree. 
 
Cn. VTir.] OF LYMPHATIC a LAXDS. 199 
 
 the conveyed cells exert a wider and more direct influence on the 
 lymphatic parenchyma; by theirangency the pattern of the primary 
 tumour is carved out of the lymphatic gland. In the least intel- 
 ligible class of cases, the influence is still more widely diffused and 
 more direct, and is indeed invisible except in its effects. But it is 
 as clearly derived from the primary tumour as in the former cases; 
 for the lymphatic gland puts on the likeness of the primary tumour 
 even to the most minute particulars. In this extreme form of in- 
 fection no transport of entire cells can be traced; but we do not 
 therefore seek for an explanation of the infection outside the 
 properties of the cell. Virchow, in treating of purulent infec- 
 tion, was led to think that the infectiveness was inherent not in 
 the cell-forms, but in the fluids that are present in or about the 
 cells; and Billroth, in speaking of cancerous infection in the passage 
 already quoted, thought that the infectiveness might reside in the 
 "serum or intercellular substance." 
 
 A more general point remains to be noticed. In the growth of 
 a secondary tumour, whether in a lymphatic gland or in an organ 
 like the liver, there is not merely a resemblance between the indi- 
 vidual tumour-cells of the secondary nodule and those of the pri- 
 mary tumour, but there is the most remarkable resemblance in the 
 arrangement of the stroma and in the general conformation of the 
 growth. The infection implies not only a transformation of the 
 cells of an infected area into the likeness of cells of the primary 
 tumour, but it implies also a consensus or co-operation among the 
 elements of the infected part to assume the grouping and general 
 plan of the primary tumour. Nothing can be more striking than 
 to find that a lymphatic gland infected from a sarcoma growing 
 from periosteum, acquires the appearance as if its own sarcomatous 
 cells had grown interstitially within, or from the germinal surface 
 of a periosteal tissue. Yet the mimicry of the infective process is 
 as complete as that. The primary tumour, if it be in an epithelial 
 organ, can be accounted for rationally by following out the func- 
 tional aberrations of the organ, and if it grow from a tissue like 
 periosteum, it can be rationally explained as a departure from the 
 cellular laws of nutrition or of growth. The ultimate form of the 
 primary tumour, one may say, is very much a matter of chance; its 
 design is not predetermined, but it is developed step by step, and 
 it may vary according to circumstances. But in the secondary 
 
200 TUMOUR-IXFECTIOX OF LYMrilATIC GLANDS. [Ft. II.] 
 
 growth there is present from tlic beginning of the process a fully 
 prepared design, which is followed as if implicitly. The primary 
 tumour is the result of natural laws and their aberrations; but 
 the secondary tumour is the result as if of a single creative fiat. 
 
 Thus, although the primary tumours of the breast are rationally 
 to be considered as nothing else than diseased breasts or parts of the 
 breast, which are still, to a certain extent, identical with or con- 
 tinuous with that organ of the body both structurally and func- 
 tionally, yet from the point of view of their infectiveness, they 
 manifest an autonomy or individuality which entitles them to be 
 considered as separate existences. But that individuality is not 
 for every tumour of the breast one and indivisible; or it should 
 rather be said that a tumour of the breast is not always one and 
 indivisible. It is a diseased process extending to more or less of 
 the mammary structure; but the issue of the diseased process is 
 not necessarily the same at all points where the process is in force. 
 In the first of the group of four cases described above, there were 
 two kinds of secondary structure found in two neighbouring lym- 
 phatic glands. One of these had the form of papillary outgrowths 
 as if from the walls of acini, and the other resembled a collection of 
 smaller acinous spaces lined with cubical epithelial cells with large 
 nuclei. Corresponding differences were found in different parts of 
 the diseased mammary tissue, as they are indeed constantl}' found 
 in the various centres of disease in the widely-extended mamma of 
 the bitch. Each of the centres can be a source of infection; and 
 the case that has been referred to seems to show that such centres, 
 existing quite close to one another, may each infect with its own 
 type of structure different l}Tnphatic glands lying in the same 
 packet. That fact does not perhaps carry the analysis of the 
 tumour-infective process much farther than it was before; but it is 
 by the establishment of many such facts that we may hope in the 
 end to analyse the process of tumour-infection, and to remove it 
 from the region of mystery. 
 
 rAMBKincE: printed by C. J. clay, M.A. at the INIVERSITY PRESS. 
 
Bedford Street, Covent Garden, London, 
 Matj, 18T7. 
 
 MACMILLAN cC- CO:S MEDICAL CATALOGUE. 
 
 Works in Physiology, Anatomy, Zoology, Botany, Chemistry, 
 Physics, Midwifery, Materia Medica, and other Professional 
 Subjects. 
 
 ALLBUTT (T. C.)— OK THE USE OF THE OPHTHALMOSCOPE 
 
 in Diseases of the Nervous System aud of tho Kidneys ; also in certain otlier 
 General Disorders. By Thomas Clifford Allbutt, M.A., M.D., Cantab., 
 Physician to the Leeds General Infirmary, Lecturer on Practical Medicine, 
 <Stc., ifec. Svo. 15s. 
 
 ANDERSON —OX THE TREATMENT OF DISEASES OF THE 
 SKIX: with an Analysis of Eleven Thousand Consecutive Cases. By Dr. 
 McCall Anderson, Professor of Clinical Medicine in the University of 
 Glasgow, and Physician to the Western Infirmary and to the Wards for Skin 
 Diseases. Crown Svo. 5s. 
 
 BASTIAN.— Works by H. Chakltox Bastiak, M.D., F.R.S., 
 
 Professor of Pathological Anatomy in University College, London, &c. : — 
 
 THE BEGINNINGS OF LIFE : Being some Account of the Nat-ire, Modes of 
 Origin, and Transformations of Lower Organisms. In Two Volumes. With 
 upwards of 100 Illustrations. Crown Svo. 'ISs. 
 
 EVOLUTION AND THE ORIGIN OF LIFE. Cro^vn Svo 6s. Gd. 
 
 ON P.\RALySIS FROM BRAIN DISEASE IN ITS COMMON FORMS. 
 
 Illustrated. Crown Svo. 10s. Gd. 
 
 " It would be a good thing if all such lectures were as clear, as systematic, 
 
 and as interesting It is of interest not only to students but to all 
 
 who make nervous diseases a study." — Journal of Mental Science. 
 
 CARTER.— A PRACTICAL TREATISE ON DISEASES OF THE 
 EYE. By R. Brudenell Carter, F.R.C.S., Ophthalmic Surgeon to St. 
 George's Hospital, &c. With Illustrations. Svo. 10s. 
 
 " No one will read Mr. Carter's book without having both his special and general 
 knowledge increased." — Lancet. 
 
 CHRISTIE.— CHOLERA EPIDEMICS IN EAST AFRICA. An 
 Account of the several Diffusions of the Disease in that country from 1821 
 till 1872, with an Outline of the Geography, Ethnology, and Trade Connec- 
 tions of the Regions thi'ough w'hich the Epidemics passed. By J. Christie, 
 M.D., late Physician to H.H. the Sultan of Zanzibar. With Maps. Svo. 15s. 
 
 COOKE (JOSIAH P., Jun.)-riRST PRINCIPLES OF CHEMICAL 
 PHILOSOPHY. By Josiah P. Cooke, Jun., Er\ine Professor of Chemistry 
 and Mineralogy in Harvard College. Third Edition, revised and corrected. 
 Crown Svo. 12s. 
 
 COOKE (M. O— HANDBOOK OF BRITISH FUNGI, with full 
 descriptions of aU the Species and Illustrations of the Genera. By M. C. 
 CooKE, M. A. Two Vols. Crown 8\ o. 24s. 
 
 2,000. 5, 77. 
 
2 MACMILLAN AND CO.'S 
 
 ELAM rc.)— A PHYSICIAN'S PROBLEMS. By Charles Elaji, 
 
 M.D., M.R.C.P. Crowa Svo. 9s. 
 
 FLOWER (W. H.)— AN INTRODUCTION TO THE OSTEOLOGY 
 OF THE M.\MMALIA. Being the substance of the Course of Lectures 
 delivered at the Royal Colle>;e of Surgeons of England in 1S70. By W. H. 
 Flower, F.R.S., F. R.C.S., Hunterian Professor of Comparative Antatomy 
 and Physiology. With numerous Illustrations. Second Edition, revised and 
 enlarged. Crown Svo. 10s. Od. 
 
 FLUCKIGER and HANBURY.— PHARMACOGRAPHIA. a 
 
 History of the Principal Drugs of Vegetable Origin met with in Great Britain 
 and British India. By F. A. Fluckiger, Ph.D., and D. Hanbury, F.R.S. 
 Svo. ISi. 
 
 " An almost faultless monograph giving a trustworthy account of 
 
 the botanical origin, history, chemical composition, source, and commercial 
 relations of the principal drugs. It abounds in research of every kind." — British 
 
 Medical Jo^irnal. 
 
 FOSTER.— A TEXT BOOK OF PHYSIOLOGY, for the tise of 
 Medical Students and others. By Michael Foster, M.D., F.R.S. Svo. 183. 
 " Dr. Foster has combined in this work the conflicting desiderata in all text- 
 books — comjirehensiveness, brevity, and clearness. After a careful perusal of the 
 whole work we can confidently recommend it, both to the student and the practi- 
 tioner as being one of the best text-books ou physiology extant."— InHCCt. 
 
 FOSTER and LANGLEY— AN ELEMENTARY COURSE OF 
 PRACTICAL PHYSIOLOGY. By Michael Foster, M.D., F.R.S., assisted 
 by J. Langley. Crown Svo. 6s. 
 
 FOSTER and BALFOUR.— ELEMENTS OF EMBRYOLOGY. 
 
 By Michael Foster, M.D., F.R.S., and F. M. B.4.LForR, M.A., Fellow of 
 Trinity College, Cambridge. With numerous Illustrations. Part I. Crown 
 Svo. 7s. fid. 
 
 " Both text and illustrations are alike remarkable for their clearness andfreedom 
 from error, indicating the immense amount of labour and care expended in the 
 production of this most valuable addition to scientific literature." — Medical Press 
 and Circular. 
 
 FOTHERGILL.— THE PRACTITIONER'S HANDBOOK OF 
 TREATMENT : or THE PRINCIPLES OF RATIONAL THERAPEUTICS. 
 By J. MiLNEi; FoTHERGiLL, M.D., M.R.C.P., Assistant Physician to the 
 Victoria Park Chest Hospital, and to the West London Hospital. Svo. 14s. 
 
 "We have every reason to thank the author for a practical and suggestive work." 
 — Lancet. 
 
 FOX.— Works by "\Yil.son Fox, M.D., Lend., F.R.C.P., F.R.S., 
 
 Holme Professor of Clinical Medicine, University College, London, Physician 
 Extraordinary to her Majesty the Queen, &:c. : — 
 
 DISEASES OF THE STOMACH: being a new and revised Edition of "The 
 Diagnosis and Treatment ok the Varieties of Dyspepsia." Svo. 8s. 6d. 
 
 ON THE ARTIFICIAL PRODUCTION OF TUBERCLE IN THE LOWER 
 ANIMALS. With Coloured Plates. 4to. 5s. 6d. 
 
 ON THE TREATMENT OF HYPERPYREXIA, as Illustrated in Acute 
 Articular Rheumatism by means of the External Application of Cold. Svo. 
 2s. Gd. 
 
MEDICAL CATALOGUE. 3 
 
 GALTON (U)— AN ADDRESS ON THE GENEEAL TRIN- 
 CIPLES WHICH SHOULD BE OBSERVED IN THE CONSTRUCTION 
 OF HOSPITALS. By Douglas Galton, C.B., F.R.S. CrowaSvo. 3s. Gd. 
 
 GRIFFITHS. -LESSONS ON PRESCRIPTIONS AND THE 
 ART OF PRESCRIBING. By W. Hansel Griffiths, Ph.D., L.R.C.P.E. 
 18mo. 3s. 6d. 
 
 " We recommend it to all students and junior members of the profession wlio 
 desire to understand the art of prescribing." — Medical Press. 
 
 HANBURY.— SCIENCE PAPERS, chiefly Pharmacological and 
 Botanical. Bv Daniel Hanbury, F.R.S. Edited with Memoir by Joseph 
 Ince, F.L.S., F.C.S. 8vo. 14s. 
 
 HOOD (Wharton.) — ON BONE-SETTING (so-called), and its 
 Relation to the Treatment of Joints Crippled by Injury, Rheumati.sm, 
 Inflammation, &c, &c. By Wharton P. Hood, M.D., M.R.C.S. Crown Svo. 
 Illustrated. 4s. 6d. 
 
 "Dr. Hood's book is full of instraction, and sho'Ud be read by all surgeons." — 
 Medical Times. 
 
 HOOKER (Dr.)— THE STUDENT'S FLORA OF THE BRITISH 
 ISLANDS. By J. D. Hooker, C.B., M.D., D.C. L.. President of the Royal 
 Society. Second Edition, revised and corrected. Globe Svo. 10s. 6d. 
 
 HUMPHRY.— Works by G. M. Humphry, M.D., r.E.S., Professor 
 of Anatomy in the University of Cambridge, and Honorary Fellow of Downing 
 
 College : — 
 
 THE HUMAN SKELETON (including the Joints). With 260 Illustrations 
 drawn from Nature. Mediiun Svo. ^Ss. 
 
 OBSERVATIONS IN MYOLOGY. Illustrated. Svo. 6s. 
 
 HUXLEY and MARTIN. — A COURSE OF PRACTICAL 
 
 INSTRUCTION IN ELEMENTARY BIOLOGY. By T. H. Huxley, LL.D. 
 Sec. B.S., assisted by H. N. Martin, M.B., D.Sc. Second Edition, revised. 
 Crown Svo. 6s. 
 
 " To intending medical students this book will prove of great value." — Lancet. 
 
 HUXLEY (Professor).— LESSONS IN ELEMENTARY PHY- 
 SIOLOGY. By T. H. Huxley, LL.D., F.R.S. With numerous Illustrations. 
 New Edition. ISmo. cloth. 4s. 6d. 
 
 LANKESTER.— COMPARATIVE LONGEVITY IN MAN AND 
 THE LOWER ANIMALS. By E. Ray Lankestek B.A. Crown Svo. 
 
 4s. 6d. 
 
 LEISHMAN.— A SYSTEJI OF MIDWIFERY, including the 
 Diseases of Pregnancy and the Puerperal State. By William Leishman, 
 M.D. , Regius Professor of Midwifery in the University of Glasgow : Physician 
 to the University Lying-in Hospital : Fellow and late Vice-President of the 
 Obstetrical Society of London, &c., tVc. Svo. Illustrated. Second and Che.iper 
 Edition. 21s. 
 
4 MACMILLAN AND CO.'S 
 
 MACLAGAN. — THE GERM THEORY APPLIED TO THE 
 
 EXPLANATION OF THE PHENOMENA OF DISEASE. By T. Maclagan, 
 
 M.D. Svo. 10s. M. 
 "We think it well that such a book as this should be written. It places before 
 the reader in clear and unmistakable language what is meant by the germ theory of 
 disease." — Lancet. 
 
 MACNAMARA.— A HISTORY OF ASIATIC CHOLERA. By 
 
 C. Macnamaba, F.C.U., Surgeon to Westminster Hospital. Crown Svo. 10s. 6d. 
 
 "A very valuable contribution to medical literature, and well worthy of the place 
 which it is sure to assume as the standard work on the subject." — Medical Examiner. 
 
 MACPHERSON.— Works by John Macpherson, !M.D. :— 
 
 THE BATHS AND WELLS OF EUROPE: their Action and Uses. With 
 Notices of Climatic Resorts and Diet Cures. With a Map. New Edition, 
 revised and enlarged. Extra fcap. Svo. 6s. 6d. 
 
 OUR BATHS AND WELLS : The Mineral Waters of the British Islands. With 
 a List of Sea-Bathing Places. Extra fcap. Svo. 3s. 6d. 
 
 MANSFIELD (C B.)— A THEORY OF SALTS. A Treatise on 
 the Constitution of Bipolar (two-membered) Chemical Compounds. By the 
 late Charles Blachford Mansfield. Crown Svo. 14s. 
 
 MAUDSLEY.— "Works by Henry Maudsley, M.D., Professor of 
 
 Medical Jurisprudence in University College, London : — 
 
 BODY AND MIND : An Inquiry into their Connection and Mutual Influence, 
 specially in reference to Mental Disorders : being the Gulstonian Lectures 
 for 1S70. Delivered before the Royal College of Physicians. New Edition, 
 with Psychological Essays added. Crown Svo. 6s. Cd. 
 
 THE PHYSIOLOGY OP MIND. Being the First Part of a Third Edition, 
 revised, enlarged, and in great part re-written, of "The Physiology and Patho- 
 logy of Mind." Crown Svo. 10s. CcJ. 
 
 THE PATHOLOGY OF MIND. [In preparation. 
 
 MIVART (St. George).— Works by St. George Mivart, F.R.S., 
 &'c.. Lecturer in Comparative Anatomy at St. Mary's Hospital : — 
 ON THE GENESIS OF SPECIES. Second Edition, to which notes have been 
 added in reference and reply to Darwin's " Descent of Man." With numerous 
 Illustrations. Crown Svo. 9s. 
 
 LESSONS IN ELEMENTARY ANATOMY. With upwards of 400 Illustrations. 
 
 ISmo. Gs. 6d. 
 "It may be questioned whether any other work on anatomy contains in like 
 compass so proportionately great a mass of information." — Lancet. 
 
 MORTON.— THE TREATMENT OF SPINA BIFIDA BY A 
 NEW METHOD. By J. Morton, M.D., Professor of Materia Mcdica, 
 Anderson's University, and Surgeon and Clinical Lecturer in the Glasgow 
 Royal Infirmary. With Illustrations. Crown Svo. 5s. 
 
 OLIVER— LESSONS IN ELEMENTARY BOTANY. By Daniel 
 
 Oliver, F.R.S., F.L.S., Professor of Botany in University College, London, 
 and Keeper of the Herbarium and Library of the Royal Gardens, Kew. With 
 nearly 200 Illustrations. New Edition. 18mo. cloth. 4s. <jd. 
 
 PETTIGREW— THE PHYSIOLOGY OF THE CIRCULATION 
 IN PLANTS, IN THE LOWER ANIMALS, AND IN MAN. By J. Bell 
 Pettiorew, M.D., F.R.S., etc. Illustrated by 150 Woodcuts. Svo. 12s. 
 
MEDICAL CATALOGUE. 5 
 
 " A more original, interesting, exhaustive, or compreliensive treatise on ttie 
 rirculation and tlie ciri'ulatory apparatus in jilants, animals, and man, lias never, 
 wo are certain, been ottered for tlie acceptance of tlio anatomist, pliysiologist or 
 student of medicine." — Vderinai'y Journal. 
 
 PIFFARD.— AN ELEMENTAKY TREATISE ON DISEASES OF 
 THE SKIN, for tlie Use of Students and Practitioners. By H. G. Piffaed, 
 JI.D., Professor of Dermatology in the University of the City of New York, 
 &c. With Illustrations. Svo. 16s. 
 
 E AD CLIFFE.— Works by Charles Bland Radcliffe, M.D., 
 F. R.C.P., Phy.sician to the Westminster Hospital, and to the National Hospital 
 for the Paralysed and Epileptic : — 
 
 DYNAMICS OF NERVE AND MUSCLE. Crown Svo. 85. U. 
 
 VITAL MOTION AS A MODE OF PHYSICAL MOTION. Crown Svo. Zs. CfJ. 
 
 RANSOME.— ON STETHOMETRY. Chest Examination by a 
 
 more Exact Method with its Results. With an Appendix on the Chemical 
 and Microscopical Examination of Respired Air. By Arthur Ransome, M.D. 
 With Illustrations. Svo. 10s. Qd. 
 
 "We can recommend his book not only to those who are interested in the 
 graphic method, but to all who are specially concerned in the treatment of diseases 
 of the chest." — British Medical Journal. 
 
 REYNOLDS (J. R.)— A SYSTEM OF MEDICINE. Vol. I. 
 Edited by J. Russell Reynolds, M.D., F.R.S. London. Second Edition. 
 Svo. 25s. 
 
 Part I. General Diseases, or Affections of the Whole System. § I. — Those 
 determined by agents operating from without, such as the exanthemata, malarial 
 diseases, and their allies. § 11. — Those determined by conditions existing within 
 the body, such as Gout, Rheumatism, Rickets, etc. Part II. Local Diseases, or 
 Affections of Particular Systems. § I. — Diseases of the Skin. 
 
 A SYSTEM OF MEDICINE. Vol. II. Second Edition. Svo. 25s. 
 
 Part II. Local Diseases (continued). § I. — Diseases of the Nervous System. 
 
 A. General Nervous Diseases. B. Partial Diseases of the Nervous System. 
 
 1. Diseases of the Head. 2. Diseases of the Spinal Column. 3. Diseases of 
 
 the Ner\'es. § II. — Diseases of the Digestive System. A. Diseases of the 
 
 Stomach. 
 
 A SYSTEM OF MEDICINE. Vol. III. Svo. 25s. 
 
 Part II. Local Diseases (continued). § II. — Diseases of the Digestive System 
 (continued), B. Diseases of the Mouth. C. Diseases of the Fauces, Pharjmx, 
 and (Esophagus. D. Diseases of the Intestines. E. Diseasesof the Peritoneum. 
 F. Diseases of the Liver. G. Diseases of the Pancreas. § III.— Diseases of 
 the Respiratory System. A. Diseases of the Larynx. B. Diseases of the 
 Thoracic Organs. 
 
 A SYSTEM OF MEDICINE. Vol. IV. Diseases of the Heart. Svo. 21.'!. 
 
 Part II. Local Diseases (continued). § IV.— Diseases of the Organs of Circula- 
 tion. A. The Heart :— Weight and Size of the Heart— Position and Form of 
 the Heart and Great Vessels — Malpositions of the Heart — Lateral or Partial 
 Aneurism of the Heart — Adventitious Products in the Heart- Pneumo-Peri- 
 cardium — Pericarditis — Adherent Pericardium — Endocarditis — Carditis— 
 Hydropericardium- Angina Pectoris and Allied States ; including certain kinds 
 of Sudden Death— Diseases of the Valves of the Heart— Atrophy of the Heart 
 — Hypertro]>hy of the Heart — Dilatation of the Heart — Fatty Diseases of the 
 Heart — Fibroid Disease of the Heart. 
 
 A SYSTEM OF MEDICINK. Vol. V. [In tlie Freu. 
 
6 MACMILLAN AND CO.'S 
 
 BICHARDSON. —DISEASES OF MODEKX LIFE. Third 
 
 Edition. Crown Svo. 12». 6d. 
 
 " The book Is in a remarkable degree original, always interesting, often singularly 
 graphic." — Times. 
 
 ON ALCOHOL. New Edition. Crown Svo. Is. 
 
 HYGEIA, A CITY OF HEALTH. Crown Svo. Is. 
 
 ROLLESTON. -THE HAEVEIAN OEATIOX, 1873. By George 
 
 KoLi.rsTON", >I.D.. F.R.S., Linacre Professor of Anatomy and Physiologj', 
 and Fellow of Mertou College, in the University of Oxford. Cro\ra Svo. 2s. 6d. 
 
 ROSCOE.— "^^orks by Henry Roscoe, F.K.S., Professor of Chemistry 
 in Owens College, Manchester : — 
 
 LESSONS IN ELEMENTARY CHEJIISTRY, INORGANIC AND ORGANIC. 
 With numerous Illustrations, and Chromolithographs of the Solar Spectrum 
 and of the Alkalies and Alkaline Earths. New Edition. ISmo. cloth, is. 6d. 
 
 CHEMICAL PROBLEMS, adapted to the above. By Professor T. E. Thorpe, 
 M.D., F.R.S.E., with Preface by Professor Roscoe. Fifth Edition, with Key. 
 ISmo. 2s. 
 
 SPECTRUM ANALYSIS. Six Lectures, with Appendices, Engravings, Maps, 
 and Chromolithographs. Tliird Edition. Royal Svo. 21s. 
 
 SCHORLEMMER.— A MAKLML OF THE CHEMISTRY OF 
 
 THE CARBON COMPOUNDS, OR, ORGANIC CHEMISTRY. By C. 
 ScHORLEMMER, F.R.S., Lecturer in Organic Chemistry in Owens College, 
 Manchester. Svo. Us. 
 
 SEATON.— A HANDBOOK OF VACCINATION. By Edward 
 C. Seaton, M.D., Medical Inspector to the Privy CoimciL Extra fcap. Svo. 
 
 Ss. 6d. 
 
 SEILER.— MICRO-PHOTOGKAPHS IN HISTOLOGY, Normal 
 and Pathological. By Cari, Seiler, M.D., in conjunction with J. Gibboks 
 Hunt, M.D., and J. G. Richardson, M.D. 
 
 This publication is intended to replace the microscope, as far as possible, for 
 those physicians who have neither opportunity nor leisure to make olsservations 
 for themselves ; and also to furnish microscopists, for comparison, correct repre- 
 sentations of t}'j)ieal specimens in the domain of normal and pathological hi.stologj'. 
 The work is issued in numbers, each containing at least four plates, with descriptive 
 letterpress ; twelve numbers to form a volume. In 4to., price 2s. 6d. each number. 
 
 SPENDER.— THERAPEUTIC MEANS FOR THE RELIEF OF 
 PAIN. Being the Prize Essay lor which the Sledical Society of London 
 awarded the Fothergillian Gold Medal in 1S74. By John Kent Spender, M.D., 
 Lond., Surgeon to the Mineral Water Hospital, Bath. Svo. 8s. 6d. 
 
 STEWART (B.)— LESSONS IN ELEMENTARY PHYSICS. By 
 
 Balfour Stewart, F.R.S., Professor of Natural Philosophy in Owens 
 Ciilk^'e, Manchester. With Numerous Illustrations and Chromolithos of the 
 Spectra of the Sun, Stars, and Nebulse. New Edition. ISmo. 4s. 6d. 
 
 WURTZ.— A HISTORY OF CHEMICAT, THEORY from the Age 
 of Lavoisier down to the present time. By Ad. Wcrtz. Translated by Henry 
 Watts, F.R.S. Crown Svo. 6s. 
 
MEDICAL CATALOGUE. 
 
 NATURE SERIES. 
 
 THE TRANSIT OF VENUS. By Professor G. Forbes. With 
 
 numerous Illustrations. Crown 8vo. 3s. Qd. 
 
 THE SPECTROSCOPE AND ITS APPLICATIONS. By J. 
 Norman Lockyek, F.K.S. With Coloured Plate and numerous Illustrations. 
 Second Edition. Crown Svo. 3s. Gd. 
 
 THE ORIGIN AND METAMORPHOSES OF INSECTS. By Sir 
 John Lubbock, M.P., F.R.S., D.C.L. With numerous Illustrations. Second 
 Edition. Crown Svo. 3s. 6d. 
 
 ON BRITISH WILD FLOWERS CONSIDERED IN RELATION 
 
 TO INSECTS. With numerous Illustrations. Second Edition. .Crown Svo. 
 4s. 6d. 
 
 THE COMMON FROG. By St. George Mivart, F.R.S. Willi 
 
 numerous Illustrations. Crown Svo. 3s. Gd. 
 
 POLARIZATION OF LIGHT. By W. Spotti-swoode, F.R.S. 
 With numerous Illustrations. Second Edition. Crown Svo. 3s. Gd. 
 
 SCIENCE PRIMERS. 
 
 Under the joint Editorship of Professors Huxley, Roscoe, and 
 Balfour Stewart. 
 
 CHEMISTRY. By H. E. Roscoe, Professor of Chemistry in Owen's 
 College, Manchester. With numerous Illustrations. Sixth Edition, with 
 Questions. ISmo. Is. 
 
 PHYSICS. By Balfour Stewart, Professor of Natural Philosophy 
 in Owens College, Manchester. Witli numerous Illustrations. Sixth 
 Edition, with Questions. ISmo. Is. 
 
 PHYSICAL GEOGRAPHY. By Archib.\ld Geikie, F.R.S., 
 Murchison Professor of Geology and Mineralogy at Edinburgh. With numerous 
 Illustrations. Seventh Edition, with Questions. ISmo. is. 
 
 GEOLOGY. By Professor Geikie, F.R.S. With numerous Illu.stra- 
 tions. Fourth Edition. ISmo. cloth. Is. 
 
 PHYSIOLOGY. By Michael Fo.ster, M.D., F.R.S. With 
 numerous Illustrations. Third Edition. ISmo. Is. 
 
 ASTRONOMY. By J. Norman Lockter, F.R.S. With numerous 
 Illustrations. Fourth Edition. ISmo. l.s. 
 
 BOTANY. By Dr. Hooker, C.B., P.R.S. With numerous 
 Illustrations. Second Edition. ISmo. Is. 
 
 LOGIC. By Professor Stanley Jevons. Second Edition. ISmo. Is. 
 
 INTRODUCTORY. By Professor Huxley. [in Preparation. 
 
8 MACMILLAN AND CO.'S MEDICAL CATALOGUE. 
 
 PRICE EIGHTEEN PENCE, MONTHLY, 
 
 THE PRACTITIONER: 
 
 |. |0uriml 0f ^Ijcraijcutks public Iji^alt^. 
 
 EDITED BY 
 
 T. LAUDER BRUNTON, M.D., F.R.S., 
 
 Fellow of the Royal College of Physicians ; 
 
 Assistant Physician to St. Bartholomcio's jflospital; and Lecturer on Materia 
 
 Medica and Therapeutics in St. Bartholomew's Hospital School. 
 
 CONTENTS, 
 
 Original Communications — Reviews of Books — Clinic of the Month— Extracts 
 from British and Foreign Journals— Notes and Queries — Bibliograishy — and the 
 Public Health Department. 
 
 Published Quarterlif, 2n-ice 6s., or Annual Suhscriiition paid in 
 Advance, 20s., post free, 21s. 
 
 THE 
 
 JOURNAL OF ANATOMY AND PHYSIOLOGY. 
 
 EDITED BY 
 
 G. M. HUMPHRY, M.D., F.R.S., Professor of Anatomy at Canibridge. 
 
 W. TURNER, MD., Professor of Anatomy at Edinhunjh. 
 
 MICHAEL P'OSTER, M.D., F.R.S., Prelector in Physiology, Trinity Colleje, 
 
 Cambridqe. 
 W. RUTHERFORD, M.D., Professor of the Institutes of Medicines, Edinburgh. 
 
 This Journal was commenced in 1SG7, and has been continued half-yearly to 
 the completion of the ninth volume. It will be continued U])on the same plan as 
 heretofore, consisting mainly of original articles in Anatomy (including Morphology, 
 Embryology, and Histology, both Vertebrate and Invertebrate), and Physiology, 
 togetlier with Reports on the Progress of Anatomy and Physiologj' and occasional 
 Notices of books. 
 
 Published every Thursday, price id. ; Monthly Parts, Is. id. and 
 Is. 8d. ; Ilalf- Yearly Volumes, 10s. M. 
 
 NATURE: 
 
 AN ILLUSTRATED JOURNAL OF SCIENCE. 
 
 Nature expounds in a popular and yet authentic manner, the Grand Results 
 OF Scientific Research, discussing the most recent scientilic discoveries, and 
 pointing out the bearing of Science upon civilisation and progress, and its claims 
 to a more general recognition, as well to a higher place in the educational system 
 of the country. 
 
 It contains original articles on all subjects within the domain of Science ; 
 Reviews setting forth the nature and value of recent Siuentific Works ; Corre- 
 spondence Columns, forming a medium of Scientific discussion and of intercommu- 
 nication among the most distinguished men of Science ; Serial Colunms, giving 
 the gist of the most important papers appearing in Scientific Journals, both 
 Homo and Foreign ; Transactions of the principal Scientilic Societies and 
 Academies of tlie World ; Notes, etc. 
 
14 DAY USE 
 
 RETURN TO DESK FROM WHICH BORROWED 
 
 E5(D)l(D)r^ Olbif airy 
 
 This book is^e on tWtast date stamped below, or 
 
 on the date to which renewed. 
 
 Renewed books are subjea to immediate recall. 
 
 Ocl7'57:^ll 
 
 m 2 4 1969 
 
 DUE 
 
 U DAYS ATTER REOft^ 
 
 INTERLIBRARY LOAa4 
 
 DEC 1968 
 
 LD 21-100m-6,'56 
 (B9311sl0)476 
 
 General Library 
 
 University of California 
 
 Berkeley