E S SAYS IN BIOLOGY HERBERT M. EVANS J^ /s- Essays in Biology In Honor of HERBERT M. EVANS WRITTEN BY HIS FRIENDS University of California Press BERKELEY AND LOS ANGELES • 1943 UNIVERSITY OF CALIFORNIA PRESS BERKELEY AND LOS ANGELES CALIFORNIA O CAMBRIDGE UNIVERSITY PRESS LONDON, ENGLAND COPYRIGHT. 1943, BY THE REGENTS OF THE UNIVERSITY OF CALIFORNIA PRINTED IN THE UNITED STATES OF AMERICA BY THE UNIVERSITY OF CALIFORNIA PRESS TO HERBERT McLEAN EVANS ON HIS SIXTIETH BIRTHDAY Y-ouR COLLEAGUES, stucleiits and other friends dedi- cate this vohime to the celebration of your anniver- sary with the confident liope tliat the truth you have so zealously sought these forty years may not be prosti- tuted or strangled in the black and desolate years that loom ahead. Your pursuit of truth and natural law has brought a fuller knowledge and understanding of fundamental problems in embryology, histology, and physiolog)'. Wis- dom, the brightest jewel in the crown of scholarship, has joined originality, insight, and energy in your labors. We look forward to watching your keen and effective pursuit of truth for many years to come. Your understanding and enthusiasm have inspired all who have worked with you and spurred them on to their utmost efforts. Your example continues to guide and stim- ulate them. We look upon you as one Of those who conquered, inch by difficult inch, The freedom of this realm of law for man; Dreamers of dreams, the buildeis of oui hope. A NOTE FROM THE PUBLICATION COMMITTEE 4 GROUP OF FRIENDS, associatcs, and students of Professor Herbert / \ McLean Evans has prepared this vokmie on the occasion of his J. V sixtieth birthday as a tribute to him in a form appropriate to his example and suggestive of his own enthusiastic endeavors. It was intended that the volume should include papers representative of the range of in- terests Avhich have characterized the last forty years of Dr. Evans's scien- tific career, namely Anatomy, Embryology, Physiology of Reproduction, Endocrinology, Nutrition, the History of Medicine, and the History of Science. Of the many scientific friends and colleagues in various parts of the world who responded to our invitation wath promises of manuscripts, some, to our regret, were prevented by the present unsettled condi- tions. To those friends of Professor Evans who were able to join with us in preparing this Sixtieth Birthday greeting, we are deeply grateful. All of us who have come in contact with Dr. Evans's genius share a common joy in suggesting to him through this book something of the appreciation we feel for his leadership, cultural ideals, and scholarly achievements. We are especially grateful to George W. Bartelmez and George \V. Corner for advice during the preparation of this volume. Publication of the book was made possible through the generous finan- cial cooperation of the following: Dr. Elmer Belt, Dr. Harold Brunn, Dr. Frederick C. Cordes, Dr. L. A. Emge, Dr. Frank Hinman, Hoffman-La Roche, Inc., Dr. William J. Kerr, Dr. Chauncey D. Leake, Dr. Herbert C. Moditt, Mr. James K. Moffitt, Schering Corporation, Dr. Francis S. Smyth, The Upjohn Com- pany, Winthrop Chemical Company. Samuel T. Farquhar, Chauncey D. Leake, William R. Lyons, Miriam E. Simpson, Cliainnaii HERBERT M. EVANS Biographical Data BORN, MODESTO, CALIFORNIA, September 23, 1882, son of Dr. C. W. and Bessie (McLean) Evans. B.S., University of California, 1904; M.D., Johns Hop- kins University, 1908; M.D. (honoris causa), Albert Ludwigs-Universitat Freiburg i.Br., 1930, and Universidad Catolica de Chile, 1941; Sc.D. (h.c), Universidad Mayor de San Marcos de Lima, 1941. Assistant, Instructor, Associate, and Associate Professor of Anatomv, Johns Hopkins University, 1908-1915; Research Associate, Carnegie Institution of W^ashington, 1913-1915; Professor of Anatomy, University of California, 1915-; Research Associate in Agriculture, 1923-25; Research Associate in Ex- perimental Biology, College of Agriculture, 1925-; Hertzstein Professor of Biology and Director of the Institute of Experimental Biology, University of California, 1930-; Honorary Member, Facultad de Biologia y Ciencias Medi- cas, Universidad de Chile, 1941; Honorary Professor, Facultad de Ciencias Medicas, Universidad Central del Ecuador, 1941. Awarded John Scott Medal, 1928. Editor: University of California Publications in Anatomy, 1921-; American Anatomical Memoirs, 1918-1939; Journal of Nutrition, 1928-1933. Fellow, American Academy of Arts and Sciences; Member, National Academy of Sci- ences; Member, Kaiserliche Leopoldinisch-Carolinische Deutsche Akademie der Naturforscher; Member, Societas Regia Medicorum Budapestinensis; Honorary Foreign Member, Academia Nacional de Medicina de Buenos Aires; Corresponding and Honorary Member, Sociedad Argentina de Biologia; Hon- orary Member, Sociedad Medica de Chile. Honorary Member, Sociedad de Biologia de Santiago de Chile. Member, American Association of Anatomists (president, 1930-1932); Member, American Medical Association; Member, American Physiological Society; Member, Society for Experimental Biology and Medicine; Fellow, American Association for the Advancement of Science (president Pacific Division, 1936-1937); Member, Association for the Study of Internal Secretions; Honorary Member, Harvey Society of New York; Honor- ary Member, Society for Biological Research, University of Pittsburgh. Mem- ber, Phi Beta Kappa; Sigma Xi; Alpha Delta Phi; Alpha Omega Alpha; Nu Sigma Nu. Member, History of Science Society, University of California. Cor- responding Member, Sociedad Argentina de Historia de la Medicina. Dele- gate, Third International Conference on Standardization of Hormones, Geneva, 1938; Delegate, Second Pan American Congress of Endocrinology. Montevideo, 1941- Faculty Research Lecturer, University of California, 1925; Bacon Lecturer. University of Illinois, 1931 ; Hertzstein Lecturer, Stanford University and Uni- versity of California, 1934; Beaumont Lecturer, Wayne County Medical So- ciety, Detroit, 1937; Jackson Lecturer, University of Minnesota, 1937; William 0^1 X Biographical Data Henry Welch Lecturer, Mount Sinai Hospital, New York, 1939; Jones Lec- turer, University of Oregon, 1940; Messenger Lecturer, Cornell University, 1942; Mellon Lecturer, University of Pittsburgh, 1942; National Sigma Xi Lecturer, 1942; Guiteras Lecturer, American Urological Association, New York, 1942. BIBLIOGRAPHY OF HERBERT McLEAN EVANS 1904-1942 K Compiled by THOMAS COWLES BIBLIOGRAPHY OF HERBERT McLEAN EVANS 1904-1942 1904 A new testraciont spine from the lower Tri- assic ol Idaho. Univ. California i'uhlns., Bull. Depmt. Geol. 3('i8):397- 402. Max, 1904. 1907 The blood-supply of lymphatic vessels in man. Anier. Jl. Anat. 7:195-208. Aug. 1, 1907. The parathyroid glandiUes. Their blood sup- ply, and their preservation in operation upon the thyroid gland. (With William S. Halsted.) Ann. Surg. 46:489-506. Oct., 1907. Notes on the resection of large portions of the small intestine. (With A. G. Brenizer.) Johns Hopkins Hosp. Bull. 18:477-480. Dec, 1907. 1908 [Book review of] "The Liver in Antiquity and the Beginnings of Anatomy," by Prof. Morris Jastrow . . . iand| ".\n Omen School Text"; . . . Anat. Recrd. 2:123-127. June, 1908. On the occurrence of newly-formed lym- phatic vessels in malignant growths. With a demonstration of their origin and ingrowth in the metasteses of a round-celled sarcoma. Johns Hopkins Hosp. Bull. 19:232-234. Aug., 1908. On an instance of two subclavian arteries of the early arm bud of man and its funda- mental significance. Anat. Recrd. 2:411- 424. Dec, 1908. 1909 On the earliest blood-vessels in the anterior limb buds of birds and their relation to the primary subclavian artery. Amer. Jl. Anat. 9:281-319. May, 1909. On the development of the aortae, cardinal and umbilical veins, and the other blood vessels of vertebrate embryos from capil- laries. Anat. Recrd. 3:498-518. Sept., 1909. 1911 (Circular request for embryological material. j [Baltimore?: 191 ij; [3i P- Die Entwicklung des BlutgefiLsssystems. In Haudbuch der Entwicklungsgcschichtc des Menschcn, bearbeitet von Charles R. Bar- deen . . . Herbert M. Evans . . . letc, etc.|. Herausgegeben von Franz Kcibel . . . und Franklin P. Mall . . . (Leipzig: 1910-1911; 2 v.) 2:551-688. 1911. 1912 I he rintcd as the preceding enirv.) Jrs.-Brcht. Schles. Gschft.f. valerland. Cult. (1913:91)1 (ii:a): 1-6. 1914. The relation between chemical (onsiitulion. physical properties, and abilitv ol ihc ben zidine dves to behave as vital stains. An.it. Recrd. 8:98-<)9. Feb. 20. 1914. The physiologv of endothelium. Anat. Recrd. 8:99-101. Feb. 20, 191.1. An experimental study of the histogenesis of the miliarv tubercle in vitally staineil rab- bits. (\Vith Fred B. Bowman Ji: M. C. Winternitz.) Jl. Exi)er. Med. 19:283-302. Mar. 1, 1914. The action of vital stains belonging lotheben- zidine group. (With ^\'erner Scluikinann.) Science (n.s.) 39:4 13-151- ^I-""- 27. 1914. An appeal to surgeons lor embryological ma terial. iBallimore?: Mav? 1911?): ill P- [xiii: XIV Bibliography Vital staining of protoplasm. Science (n.s.) 39: 843-844. June 5, 1914. Die vitale Tarbung mit'sauren Farbstoffen in ihrer Bedeutung fiir pharmakologische Probleme. Ein Beitrag zur Pharmakologie kolloider Losungen. (With Werner Schule- mann.) Mit einem kolloid-chemischen Bei- trac:. Von Felix Wilborn. Deutsch. med. Wchnschr. 40:1508-1511. July 23, 1914. 1915 Ober Natur mid Genese der durch saure Farbstolfe entstehenden Vitalfarbungsgra- nula. (With Werner Schulemann.) Fol. haematol., Arch. 19:207-19. Feb. 25, 1915. The macrophages of mammals. Amer. Jl. Physiol. 37:243-258. May 1, 1915. 1916 On the behavior of the ovary and especially of the atretic follicle towards vital stains of the azo group. Anat. Recrd. 10:264. Jan. 20, 1916. On the behavior of the mammalian ovary and especially of the atretic follicle towards vital stains of the acid azo group. Proc. Soc. Exper. Biol. & Med. 13:80-81. [Mar.], 1916. 1917 A human embryo of seven to eight somites. (With G. W. Bartelmez.) Anat. Recrd. 11: 355-356. Jan. 20, 1917. 1919 On the segregation of macrophage and fibro- blast cells by means of vital acid dyes and on the cause of the differential effect of these substances. (With Katharine J. Scott.) Anat. Recrd. 16:148-149. May 20, 1919. 1920 Theoestrous cycle in the rat and other studies in the physiology of reproduction. (With Joseph A. Long.) ([PPreprint of the next but one and the following entries for the year; dated:j March, 1920); 11 p. Blood volume studies. III. Behavior of large series of dyes introduced into the circulat- ing blood. (With A. B. Dawson & G. H. Whipple.) Amer. Jl. Physiol. 51:232-256. Mar. 1, 1920. The oestrus cycle in the rat. (With Joseph A. Long.) Anat. Recrd. 18:241-244. Apr. 20, 1920. On the attainment of sexual maturity and the character of the first oestrus cycle in the rat. (With Joseph A. Long.) Anat. Recrd. 18:244-245. Apr. 20, 1920. Effect on the oestrus cycle of the removal of various portions of the reproductive sys- tem. (With Joseph A. Long.) Anat. Recrd. 18:245. Apr. 20, 1920. Rhythmical recurrence of the typical oestrus cycle after ovarian transplantation. (With Joseph A. Long.) Anat. Recrd. 18:245-246. Apr. 20, 1920. The effect of copulation in delaying the oc- currence of the next oestrus cycle and the production of a similar effect by mechan- ical stimulation of the cervix. (With Joseph A. Long.) Anat. Recrd. 18:246-247. Apr. 20, 1920. The inhibition of oestrus and ovulation by lactation. (With Joseph A. Long.) Anat. Recrd. 18:247. Apr. 20, 1920. Corpora lutea of lactation in the rat as dis- tinguished from the corpora of pregnancy and those of ovulation. (With Joseph A. Long.) Anat. Recrd. 18:247-248. Apr. 20, 1920. The survival and time of disappearance of the corpora lutea of pregnancy in the rat under \'arious conditions. (With Joseph A. Long.) Anat. Recrd. 18:248. Apr. 20, 1920. The experimental production of deciduo- mata in the rat, with special reference to the phases of the oestrus cycle. (With Joseph A. Long.) Anat. Recrd. 18:248-249. Apr. 20, 1920. A characteristic sign of pregnancy in the rat detectable from the thirteenth to the six- teenth day. (With Joseph A. Long.) Anat. Recrd. 18:249. Apr. 20, 1920. 1921 On the differential reaction to vital dyes ex- hibited by the two great groups of con- nective-tissue cells. (With Katharine J. Scott.) Contribs. Embryol., Carnegie Instn. Washington io(47):i-55. [I921.] Further studies in the physiology of repro- duction. (With Joseph A. Long.) (i?Pre- print of the next ten entries, together with three other abstracts; dated:] March, 1921); 10 p. Proportion of ova producing full-term young in the rat. (With Joseph A. Long.) Anat. Recrd. 21:56-57. Apr. 20, 1921. On the production of the condition of 'pseudopregnancy' by infertile coitus or mechanical stimulation of the cervical canal in the rat. (With Joseph A. Long.) Anat. Recrd. 21:57. Apr. 20, 1921. A characteristic histology of the vaginal mu- cosa during lactation. (With Joseph A. Long.) Anat. Recrd. 21:58. Apr. 20, 1921. Bibliography XV On the production of dcciduomata during lactation. (With Joseph A. Long.) Anal. Recrd. 21:58-59. Apr. 20, 1921. On the rapid maturation of the ovary by transplantation of the youthful gonad to adults. (With Joseph A. Long.) Anat. Recrd. 21:60. Apr. 20, 1921. On the association of continued cornification of the vaginal mucosa \vith the presence of large vesicles in the ovarv and the absence of corpus formation. (With Joseph A. Long.) Anat. Recrd. 21:60-61. Apr. 20,1921. The effect of thyroid feeding on the oestrous cycle of the rat. (With Joseph A. Long.) Anat. Recrd. 21:61. Apr. 20, 1921. The effect of thyroidectomy on the oestrous cycle of the rat. (With Joseph A. Long.) Anat. Recrd. 21:61-62. Apr. 20, 1921. The effect of feeding the anterior lobe of the hypophysis on the oestrous cycle of the rat. (^Vith Joseph A. Long.) Anat. Recrd. 21:62. Apr. 20, 1921. 1 he effect of the anterior lobe administered intraperitoneally upon growth, maturity, and oestrous cycles of the rat. (With Joseph A. Long.) Anat. Recrd. 21:62-63. Apr. 20, 1921. Relation of nutrition to the oestrous cycle. (With Katharine J. Scott.) Anat. Recrd. 21:80-81. Apr. 20, 1921. 1922 1 he rhythm of gonadal function with special reference to the relations between uterus and ovary. In Endocrinology and Metabo- lism . . . edited by Lewellys F. Barker . . . R. G. Hoskins . . . [andj Herman O. Mosen- thal. (New York, London: 1922; 5 v.) 2:572- 599- On an invariable and characteristic disturb- ance of reproductive function in animals reared on a diet poor in fat soluble vita- mine A. (With Katherine Scott Bishop.) Anat. Recrd. 23:17-18. Jan. 20, 1922. Oestrus and ovulation in the rat, from a large group of animals on a standard diet. (W^ith Katherine Scott Bishop.) Anat. Recrd. 23: 18-19. Jan. 20, 1922. Characteristic effects upon growth, oestrus, and ovulation induced by the intraperi- toneal administration of fresh anterior hypophyseal substance. (With Joseph A. Long.) Anat. Recrd. 23:19. Jan. 20, 1922. Absence of monocytes in thoracic duct lymph. (With George \V. Thorne.) Anat. Recrd. 23:42. Jan. 20, 1922. On the relations between fertility and nutri- tion. L The ovulation rhythm in the rat on a standard nulritional regime. (With Katherine S. Bishop.) Jl. Metabol. Resrch. ':3' 9-333- I'e'j-. 1922. On the relations between fertility and nutri- tion. II. — The ovulation rhvthm in the rat on inadecjuate nutritional regimes. (With Katharine Scott Bishop.) Jl. Metabol. Resrch. 1:335-356. Feb., 1922. Characteristic effects upon growth, oestrus and ovulation induced by the intraperi- toneal administration of fresh anterior hypophyseal substance. (^Vith J. .\. Long.) Proc. Natn. Acad. Scis. 8:38-39. Mar. [15], 1922. The oestrous cycle in the rat and its asso- ciated phenomena. (With Joseph A. Long.) Mems. Univ. California 6: [2j, 148 p. June 28, 1922. On the existence of a hitherto unrecognized dietary factor essential for reproduction. (With K. Scott Bishop.) Science (n.s.) 56: 650-651. Dec. 8, 1922. 1923 On the relations between fcrtilitv and nu- trition. III. The normal reproductive per- formance of the rat. (With Katharine Scott Bishop.) Jl. Metabol. Resrch. 3:201-231. Feb., 1923. [On the relations between fertility and nu- trition. IV.] The production of sterility with nutritional regimes adequate for gro\\th and its cure with other foodstuffs. (^Vith Katharine Scott Bishop.) Jl. Metabol. Resrch. 3:233-316. Feb., 1923. On the existence of a hitherto unknown dietary factor essential for reproduction. (With K. Scott Bishop.) Amer. Jl. Physiol. 63:396-397- Fet)- 1. 19-3- Separation of the principle in the anterior hypophysis affecting ovulation from that controlling general body growth. (Willi C. F. Flower, C. E. Forkner, \\\ E. KcUum, A. T. Walker & P. E. Smith.) Anat. Recrd. 25:107. Apr. 20, 1923. Stability of hormones in the anterior hv- pophvsis. (\\'ith C. F. Flower, C. E. Fork- ner, W. E. Kellum, A. T. Walker j^: P. E. Smith.) Anat. Recrd. 25:107. Apr. 20, 1923. Participation of the mammary gland in the changes of pscudopregnancy in the rat. (With Mary E. Freyer.) Anat. Recrd. 25: 108. .\pr. 20, 1923. Independence of the oestrous or heat hor- mone from that causing 'oestrous' struc- tural changes in tiic vaginal mucosa of the rat. (With K. Scott Bishop.) Anat. Recrd. 25:128. Apr. 20, 1923. XVI Bibliography The cause of reproductive upset in dietary deficiencies due to lack of vitamine A. (Witli K. Scott Bisliop.) Anat. Recrd. 25: 129. Apr. 20, 1923. Existence of a hitlierto unknown dietary fac- tor essential for reproduction. (With Kath- arine S. Bishop.) Jl. Amer. Med. Assn. 81: 889-892. Sept. 15, 1923. 1924 The rhythm of gonadal function with special reference to the relations between uterus and ovary. (Reissue of the first entry in 1922.) In Endocrinology and Metabolism . . . edited by Lewellys F. Barker . . . R. G. Hoskins . . . ^andj Herman O. Mosenthal. (Reissue; New York, London: i924([Ci922]); 5 V.) 2:572-599. On the existence of a hitherto unknown dietary factor essential for reproduction. Proc. World's Dair. Congr. (1923) 2:1027- 1034. 1924. Preliminary steps in the isolation and con- centration of vitamine X. (With George O. Burr.) Anat. Recrd. 27:203. Apr. 25, 1924. Stability and solubilities of the food sub- stance or vitamine X required for repro- duction. (With Katherine S. Bishop.) Anat. Recrd. 27:203-204. Apr. 25, 1924. Proof of the power of the body to store the substance X recjuired for reproduction. (With Katherine S. Bishop.) Anat. Recrd. 27:204. Apr. 25, 1924. Unique dietary needs for lactation. Science (n.s.) 60:20-22. July 4, 1924. 1925 The function of the anterior hypophysis. Harvey Lects. (1923-1924) 19:212-235. 11925-1 The relations between fertility and nutri- tion. In Lectures on Nutrition: a series . . . given at the Mayo Foundation and the universities of Wisconsin, Minnesota, (etc., etc.] . . . 1924-1925. (Philadelphia and Lon- don: [C1925]) 209-234. Preparation of Brilliant Congo R ("Vital Red") and the suitability of various sam- ples of Vital Red for blood volume work. (With Samuel Palkin.) Jl. Amer. Chem. Soc. 47:429-434. Feb. 15], 1925. Characteristics of the anterior hypophyseal hormones. (With Miriam E. Simpson.) Anat. Recrd. 29:356. Mar. 25. 1925. Distribution of vitamine X in natural foods. (With George O. Burr. ) Anat. Recrd. 29: 356. Mar. 25, 1925. The repair of dwarfism following thyroidec- tomy by the administration of anterior hypophyseal fluid. (With Charles F. Flower.) Anat. Recrd. 29:383. Mar. 25, 1925. The anti-sterility vitamine fat soluble E. (With George O. Burr.) Proc. Natn. Acad. Scis. 11:334-341. June [15], 1925. Invariable occurrence of male sterility with dietaries lacking fat soluble vitamine E. Proc. Natn. Acad. Scis. 11:373-377. J'^^'Y [151. 1925- c 1926 Development of the human embryo during the period of somite formation, including embroys \vith 2 to 16 pairs of somites. (With George W. Bartelmez.) Contribs. Embryol., Carnegie Instn. Washington 17(85): 1-67. Feb., 1926. Effects of anterior hypophyseal extracts on the male. (With Miriam E. Simpson.) Anat. Recrd. 32:206. Mar. 25, 1926. The initial infertility of early copulations in the normal male rat. Anat. Recrd. 32:206. Mar. 25, 1926. Increased efficacy of subcutaneous when com- pared with intraperitoneal administration of the ovarian hormone. (With George O. Burr.) Amer. Jl. Physiol. 77:518-521. Aug. 1, 1926. 1927 .\ chronological list of the chief discoveries relating to vitamines. Compend. Med. & Surg. 5:9-11. 20. Jan., 1927. The oestrous cycle in the dog. I. The vaginal smear. (With Harold H. Cole.) Anat. Recrd. 35:10. Mar. 25, 1927. The oestrous cycle in the dog. II. Coincident changes in the genital organs. (With Harold H. Cole.) Anat. Recrd. 35:11. Mar. 25, 1927. A new epithelial sensory organ in the va- ginal mucosa of the dog. Anat. Recrd. 35: 36. Mar. 25, 1927. Resemblance of litter-mates in the degree of response to anterior hypophyseal fluid. (With Miriam E. Simpson.) Anat. Recrd. 35:36. Mar. 25, 1927. Experimental gigantism — differential effect of anterior hypophyseal extract on normal and oonadectomized males and females. (With Miriam E. Simpson.) Anat. Recrd. 35:36-37- Mar. 25, 1927. A new dietary deficiency with highly puri- fied diets. (With George O. Burr.) Proc. Soc. Exper. Biol. & Med. 24:740-743. May, 1927. The synthesis of \itamin E by plants grown in culture solutions. (With Dennis R. Bibliography XVI 1 Hoagland.) Amcr. Jl. Physiol. 80:702-704. May 1, 1927. Vitamin E: the ineffectiveness of curative dosage when mixed with cHets containing high proportions of certain fats. (With George O. Burr.) Jl. Amer. Med. Assn. 88: 1462-1465. May 7, 1927. A new dietary deficiency produced with highly purified diets. (With George O. Burr.) Jl. Biol. Chein. 74:lxxii-lxxiii. July, 1927- The antisterility vitamine: fat soluble E. (With George O. Burr & Theodore L. Althausen.) Mems. Univ. California 8: [7], 176 p. Aug., 1927. New dietary deficiency with highly purified diets. II. Supplementary requirement of diet of pure casein, sucrose, and salt. (With George O. Burr.) Pioc. Soc. Exper. Biol. & Med. 25:41-48. Oct., 1927. \'itamin E. IL The destructive efTect of cer- tain fats and fractions thereof on the anti- sterility vitamin in wheat germ and in wheat germ oil. (With George O. Burr.) Jl. Amer. Med. Assn. 89:1587-1590. Nov. 5, 1927. ^ ' 1928 On the amount of \ itamin B required during lactation. (With George O. Burr.) Jl. Biol. Chem. 76:263-272. Jan., 1928. Development of paralysis in the suckling young of mothers deprived of \itamin E. (With George O. Burr.) Jl. Biol. Chem. 76:273-297. Jan., 1928. A new dietary deficiency with highlv purified diets. III. The beneficial effect of fat in the diet. (With George O. Burr.) Proc. Soc. Exper. Biol. & Med. 25:390-397. Feb., 1928. \'itamine E and anaemia. (With Clara L. Rohls.) Anat. Recrd. 38:52. Mar. 25. 1928. A ne\v differentiation between the antineu- ritic \ itamin B and the purely growth-pro- moting \ itamin B. (With George O. Burr.) Jl. Biol. Chem. 77:231-240. Apr., 1928. The effects of inadequate vitamin A on the sexual physiology of the female. Jl. Biol. Chem. 77:651-654. May, 1928. Spontaneous deciduomata in pseudopreg- nancy with lo\v vitamin E. Amer. Jl. Physiol. 85:149-153. May 1. 1928. Sterility in inbred rats. Amer. Jl. l^hysiol. 85:154-157. May 1, 1928. La vitamine lipo-soluble E ct le reproduction chez les Mammiferes. Bull. Soc. sci. d'hyg. aliment. 16:382-397. 1928. The efTect of inadequate \itamin B upon sexual jjhysiology in the male. Jl. Xutrit. 1:1-21. Sept., 1928. Relation of \itaiiiiii )■, to growth and vigor. Jl. Nuliii. 1:23-28, Sept., 1928. Sparing action of fat on the anti-neinitic vitamin. (With Samuel Lepkovsky.j Science (n.s.) 68:298. Sept. 28, 1928. The oestroiis cvcle in the dog. I. The vaginal smear. (With Harold H. Cole.) (Republica- tion of paper issued in 1927.) Cornell X'elerinin. 18:352. Oct., 1928. The oestrous cycle in the dog. II. Coincident changes in the genital organs. (With Harold H. Cole.) (Republication of paper issued in 1927.) Cornell Velerinrn. 18:357. Oct., 1928. A new cpitheli;il sensory organ in the vaginal mucosa of the dog. (Republication of paper issued in 1927.) Cornell Veterinrn. 18:384. Oct., 1928. A sex cHllerence in chromosome lengths in the mammalia. (With Olive Swezy.) Ge- netics 13:532-513- Nov., 1928. Antagonism of growth and sex hormones of the anterior hypophysis. (With Miriam E. Simpson.) Jl. Amer. Med. Assn. 91:1337- 1338. Nov. 3, 1928. Occurrence of faint bleeding on a definite intermenstrual day in man. (With Miriam E. Simpson.) Science (n.s.) 68:453. ^^«v. 9. 1928. 1929 Comparison of anterior-hypophyseal im- plants from normal and gonatlectomized animals, with reference to their capacity to stimulate the immature o\ary. (With Miriam E. Simpson.) .Vnat. Recrd. 42:48. Mar. 25, 1929. Comjjarison of anterior-hvpophvseal im- plants from the male and female, with ref- erence to their capacity to stimulate the immature ovarv. (With Miriam E. Simp- son.) Anat. Recrd. 42:48. Mar. 25, 1929. (Book re\ iew of] A Short History of Medicine. Bv Charles Singer. Oxford University Press, New ^■olk, 1928 . . . Univ. California Chroncl. 31:211-214. Apr., 1929. [Book review ofj An Introduction to the His- tory of Medicine. By Eielding H. Garrison. Edition 4. Philadelphia. W. B. Saunders. u)2() . . . Univ. California Chroncl. 31:21 )- 216. A|)r., 1929. Impairment of the hirtii mechanism clue 10 hormones from the anterior Inpophysis. (With Miriam E. Simpson.) Proc. Soc. Exper. Biol. & Med. 26:595-597. .\pr., 1929- Stimulation of placenloma reaction in vir- ginal endometrium bv irealment with an- terior hvpophyseal hormone. iWith Mir- XVI 11 Bibliography iani E. Simpson.) Proc. Soc. Exper. Biol. & Med. 26:597. Apr., 1929. Hyperplasia of mammary apparatus in pre- cocious maturity induced by anterior hy- pophyseal hormone. (With Miriam E. Simpson.) Proc. Soc. Exper. Biol. & Med. 26:597-598- Apr., 1929. Hyperplasia of mammary apparatus of adult virginal females induced by anterior hy- pophyseal hormones. (With Miriam E. Simpson.) Proc. Soc. Exper. Biol. & Med. 26:598. Apr., 1929. A comparison of anterior hypophyseal im- plants from normal and gonadectomized animals with reference to their capacity to stimulate the immatine ovary. (With Mir- iam E. Simpson.) Amer. Jl. Physiol. 89:371- 374. July 1, 1929. A sex difference in the hormone content of the anterior hypophysis of the rat. (With Miriam E. Simpson.) Amer. Jl. Physiol. 89: 375-378- July 1. 1929- The effect of pregnancy on the anterior hy- pophysis of the rat and co^v as judged by the capacity of implants to produce pre- cocious maturity. (With Miriam E. Simp- son.) Amer. Jl. Physiol. 89:379-380. July 1, 1929- A comparison of the ovarian changes pro- duced in immature animals by implants of hypophyseal tissue and hormone from the urine of pregnant Avomen. (With Mir- iam E. Simpson.) Amer. Jl. Physiol. 89:381- 387. July 1, 1929. Sparing action of fat on the antineuritic vitamin B. (With Samuel Lepkovsky.) Jl. Biol. Chem. 83:269-287. Aug., 1929. Technique for determination of the antineu- ritic vitamin B. (With Samuel Lepkovsky.) Jl. Nutrit. 2:1-5. Sept., 1929. The chromosomes in man: sex and somatic. (With Olive Swezy.) Mems. Univ. Califor- nia 9(i):[V], 1-64. Oct., 1929. Maturation of human embryonic ova. (With Olive Swezy.) Proc. Soc. Exper. Biol. & Med. 27:10. Oct., 1929. Ovogenesis in the mammalia. (With Olive Swezy.) Proc. Soc. Exper. Biol. & Med. 27: 11. Oct., 1929. On some relations of vitamin B to fat and carbohydrate metabolism. (With Samuel Lepkovsky.) Amer. Jl. Physiol. 90:340. Oct. 1, 1929. Potent, sterile and low-protein extracts of the growth hormone from the anterior hy- pothesis. (With Robert E. Cornish Sc Mir- iam E. Simpson.) Proc. Soc. Exper. Biol. & Med. 27:101-102. Nov., 1929. 193" Ovarian changes during pregnancy in the rat. (With Olive Swezy.) Science (n.s.) 71:46. Jan. 10, 1930. Aschheim-Zondek test for pregnancy — its present status. (With Miriam E. Simpson.) California & West. Med. 32:145-148. Mar., i93'>- The effect of ions on sedimentation of col- loidal particles by means of the centrifuge. (With R. E. Cornish.) Jl. Amer. Chem. Soc. 52:1009-1012. Mar. ;6j, 1930. Different effects secured from intraperitoneal as contrasted with subcutaneous adminis- tration of the anterior-hypophyseal hor- mones. (With Miriam E. Simpson.) Anat. Recrd. 45:215. Apr. 25, 1930. Some effects on the hypophysis of hyper- and hypothyroidism. (With Miriam E. Simpson.) Anat. Recrd. 45:215. Apr. 25, 1930- Subnormal sex-hormone content of the hy- pophysis of animals with inadequate anti- neuritic vitamin B. (With Miriam E. Simp- son.) Anat. Recrd. 45:216. Apr. 25, 1930. Glucose tolerance in avitaminosis due to low antineuritic vitamin B. (With Samuel Lep- kovsky & Clarence Wood.) Jl. Biol. Chem. 87:239-250- June. 1930- The human ovarian germ cells. (With Olive Swezy.) Jl. Morphol. & Physiol. 49:543-577- June 5, 1930. Construction and use of Raschig's laboratory fractionating column. (With R. E. Cornish, S. Lepkovsky, R. C. Archibald &: G. Feskov.) Indust. & Engin. Chem., Analyt. Ed. 2:339- 343. July 15, 1930. A condenser for low temperature evapora- tion of water. (With R. E. Cornish & J. C. Atkinson.) Jl. Amer. Chem. Soc. 52:4334- 4335- Nov. 15], 1930. 1931 On the preparation of a concentrated source of the heat-labile vitamin B, free from con- tamination with the heat-stabile factor G. (With Samuel Lepkovsky.) Jl. Nutrit. 3: 353-374- Jan., 1931. Sex differences in the response of the rat to hypophyseal growth hormone. (With Mir- iam E. Simpson.) Anat. Recrd. 48(supp.): 18. Feb. 25, 1931. Anemia and rickets. (With Clara L. Kohls.) Anat. Recrd. 48(supp.):43. Feb. 25, 1931- Stimulating effect of hypophyseal implants in male rats with inadequate vitamine G. Bibliography \i\ (With Miriam E. Simpson.) Anat. Rocid. 48(supp.):43- Feb. 25, 1931. Treatment of the congenital dwar[ Aviih growth hormone (rat). (With Miriam E. Simpson.) Anat. Rccrd. 48(supp.):43. Feb. 25' 1931- Necessary concmrente of o\'ary in mammary response to the hypophysis in the rat. (^Vith Miriam E. Simpson.) Anat. Recrd. 48 (siipp.):44. Feb. 25, 1931. Tire uterus-ovary relationship and its l)car- ing on the time of ovulation in primates. (With Olive Swezy.) Amer. Jl. Physiol. 96: 628-639. Mar. 1, 1931. Ineffectiveness of prolan in hypophysccto- mized animals. (With Frederick L. Reich- ert, Richard I. Pencharz, Miriam E. Simp- son & Karl Meyer.) Proc. Soc. Expcr. Biol. & Med. 28:843-844. May, 1931. Relation of prolan to anterior h\pophyseal hormones. (With Karl Meyer & Miriam E. Simpson.) Proc. Soc. Exper. Biol. S: Med. 28:845-847. May, 1931. Beneficial effects of fat in high sucrose diets when the requirements for antineuritic vitamin B and the fat-soluble vitamin are fully satisfied. (With Samuel Lepkovsky.) Jl. Biol. Cheni. 92:615-622. Aug., 1931. The effects of hypoph}seal hormones on ovo- genesis in the foetal ovary. (\Vith Olive Swezy.) Anat. Recrd. 50:189-192. Aug. 25. 1931- Effect of adrenalectomy on the testes of the rat. (With S. C. Freed & B. Brownfield.) Proc. Soc. Exper. Biol. & Med. 29: 1-3. Oct., 1931- Hormones of the anterior hypophysis. (\Vith Miriam E. Simpson.) Amer. Jl. Physiol. 98: 511-546. Oct. 1, 1931. Researches on the relation of gaseous metab- olism of animals to endocrine and vitamin deficiency. Carnegie Instn. AVashington Yr. Bk. (1930-1931) 30:461. Dec. 10, 1931. An introduction to the study of the oestrous cycle in the dog. (\Vith Harold H. Cole.) Mems. Univ. California 9(2):vii, 65-118. Dec. 19, 1931. Ovogenesis and the normal follicular cycle in aciult mammalia. (With Olive Swezy.) Mems. Univ. California 9(3):[^'i]' n9-224- Dec. iQ, 1931. 1932 Ovogenesis and the normal follicular cycle in adult mammalia. (With Olive Swezy.) California & West. Med. 36:60. Jan., 1932. Testicular degeneration due to inadequate vitamin A in cases where E is adccpiate. Amer. Jl. Physiol. 99:477-486. Jan. 1, 1932. Relation of prolan to ilic aiucrior hypo- physeal hormones. (With Karl Meyer & .Miriam E.Simpson.) Amer. Jl. Physiol. 100: 141-156. Mar. 1, 1932. Relative ineffectiveness of prolan in hypopli- ysectomizcd animals. (With F. L. Reichert. R. I. Pencharz, .\I. E. Simpson & K. Meyer.) .\nier. Jl. Physiol. 100:157-161. Mar. 1. '932- Disturbance of carbohydrate metabolism in normal dogs injected with the hypophyseal growth hormone. (With Karl Meyer, Mir- iam E. Simpson & Frederick L. Reichert.) Proc. Soc. Exper. Biol. & Med. 29:857-858. Apr., 1932. Vital need of the body for certain unsaturated fatty acids. I. Experiments with fat-free diets in which sucrose furnishes the sole source of energy. (With Samuel Lepko\ ■ sky.) Jl. Biol. Chcm. 96:143-156. Apr., 1932- \'ital need of the body for certain unsaturated fatty acids. II. Experiments with high fat ' diets in which saturated fatty acids furnish the sole source of energy. (\Vith Samuel Lepkovsky.) Jl. Biol. Chem. 96:157-164. Apr., 1932. The sparing action of fat on vitamin B. 11. The role played by the melting point and the degree of unsaturation of various fats. (With Samuel Lepkovsky.) Jl. Biol. Chem. 96:165-177. Apr., 1932. The sparing action of fat on vitamin B. IH. The role played by glycerides of single fatty acids. (With Samuel Lepkovsky.) Jl. Biol. Chem. 96:179-188. Apr., 1932. Cure of the cachexia following hypoi^hysec- tomy by administration of the growth hor- mone and its relation to the resulting adreno-cortical repair. (With Karl Meyer, Richard Pencharz & Miriam E. Simp- son.) Science (n.s.) 75:442--i43. Apr. 22, 1932- Vitamin E. Jl. .Vmer. Med. .\ssn. 99:409-475- ,\ug. 6, 1932. Ovogenesis and the hypophysis. (With Olive Swezy.) Anat. Recrd. 54(supp.):48. Nov. 25, 1932. Vital need of the body for certain unsaturated fatly acids. IIL Inability of the rat organ- ism to synthesize the essential unsaturated fatty acids. (With Samuel Lepkovsky.) Jl. Biol. Chem. 99:231-234. Dec, 1932. The sparing action of fat on vitamin B. 1\'. Is it necessary for fat to interact with \itainin B in the alimentarv canal to exert its sparing elfeci? (With Samuel Lepko\ - sky.) Jl. Biol. Chem. 99:235-236. Dec, 1932. XX Bibliography The sparing action of fat on vitamin B. V. The role of glycerides of oleic acid. (With Samuel Lepkovsky.) Jl. Biol. Chem. 99:237-240. Dec, 1932. 1933 The hypophyseal substance giving increased gonadotropic effects ^vhen combined with prolan. (With Miriam E. Simpson &: Paul R. Austin.) Jl. Exper. Med. 57:897-906. June 1, 1933. The growth and gonad-stimulating hormones of the anterior hypophysis. (With Karl Meyer, Miriam E. Simpson, Alexander J. Szarka, Richard I. Pencharz, Robert E. Cornish & Frederick L. Reichert.) Mems. Univ. California ii:[xiij, 446 p. Aug. 2, 1933- Present position of our knowledge of ante- rior pituitary function. Jl. Amer. Med. Assn. 101:425-432. Aug. 5, 1933. Peculiarities of the prolan-like substance in urine in a case of embryonal carcinoma of the testis. (With Miriam E. Simpson, Paul R. Austin & Russell S. Ferguson.) Pioc. Soc. Exper. Biol. & Med. 31:21-23. Oct., 1933. Further studies on the hypophyseal substance gi\ ing increased gonadotropic effects when combined with prolan. (With Miriam E. Simpson & Paul R. Austin.) Jl. Exper. Med. 58:545-559- Nov. 1, 1933. The recognition and comparison of prolan and prolan-like substances. (With Miriam E. Simpson & Paul R. Austin.) Jl. Exper. Med. 58:561-568. Nov. 1, 1933. Concentration of the gonadotropic hormone in pregnant mare's serum. (With Edwin L. Gustus & Miriam E. Simpson.) Jl. Exper. Med. 58:569-574. Nov. 1, 1933. The female sex hormones: the present status of our knowledge of anterior pituitary function. Trns. Congr. Amer. Physns. & Surgns. (1933) 15:25-46. 1933. 1934 Hypertrophy of the female pituitary follow- ing injection of gonadotropic hormone. (With Miriam E. Simpson &: Morvyth Mc- Queen-Williams.) Univ. California Publns. Anat. i(5):|ii), 161-165. Jan. 27, 1934. Hypophyseal infantilism; treatment with an anterior hypophyseal extract: preliminary study. (With E. Kost Shelton & Lyman A. Cavanaugh.) (PPreprint of third following entry.) Trns. Sect. Pediat., Amer. Med. Assn. 1933:235-255. [?Mar., 1934., Prevention of [i.e., byj ovariotomy of the hy- pophyseal hypertrophy following chronic administration of gonadotropic hormone. (W^ith Miriam E. Simpson.) Anat. Recrd. 58(abstr. supp.):62. Mar. 25, 1934. Scrotal hypertrophy following chronic in- jection of gonadotropic hormone. (With Miriam E. Simpson.) Anat. Recrd. 58 (abstr. supp.):62. Mar. 25, 1934. Hypophyseal infantilism; treatment with an anterior liypophyseal extract: preliminary study. (With E. Kost Shelton & Lyman A. Cavanaugh.) (PReprint of third preceding entry.) Amer. Jl. Diseases Childr. 47:719- 736. Apr., 1934. Purification of vitamins: fractional distribu- tion between immiscible solvents. (With R. E. Cornish, R. C. Archibald 8: Elizabeth A. Murphy.) Indust. & Engin. Chem., [Indust. Ed.], 26:397-406. Apr., 1934. The sparing action of fat on vitamin B.(With Samuel Lepkovsky.) Jl. Biol. Chem. 105: xxvii. May, 1934. First editions in the history of science. In Exhibition of first editions of epochal achievements in the history of science. Sponsored by a committee of the History of Science Club of the University of Cali- fornia . . . (Berkeley: Jime, 1934) 3-6. On a selective gametogenic effect of certain hypophyseal extracts. (With Richard I. Pencharz & Miriam E. Simpson.) Science (n.s.) 80:144. Aug. 10, 1934. Vital need of the body for certain unsaturated fatty acids. IV. Reproduction and lactation upon fat-free diets. (With Samuel Lep- kovsky & Elizabeth A. Murphy.) Jl. Biol. Chem. 106:431-440. Sept., 1934. \'ital need of the body for certain unsatu- rated fatty acids. V. Reproduction and lactation upon diets containing satu- rated fatty acids as their sole source of energy. (\Vith Samuel Lepkovsky & Eliza- beth A. Murphy.) Jl. Biol. Chem. 106:441- 444. Sept., 1934. \'ital need of the body for certain unsatu- rated fatty acids. VI. Male sterility on fat- free diets. (With Samuel Lepkovsky S: Elizabeth A. Murphy.) Jl. Biol. Chem. 106: 445-450. Sept., 1934. The repair of the reproductive system of hypophyscctomized female rats by com- binations of an hypophyseal extract (syner- gist) with pregnancy-prolan.(With Richard I. Pencharz & Miriam E. Simpson.) Endo- crinology 18:601-606. Sept./Oct., 1934. Maintenance and repair of the reproductive system of hypophyscctomized male rats by hypophyseal synergist, pregnancy-prolan and combinations thereof. (With Richard Bibliography XXI I. Pencharz &: Miriam E. Simpson.) Endo- crinology 18:617-618. Sept./Ott., 1931. The sparing action of fat on vitamin B. VI. The influence of the lc\els of |)rotcin and vitamin G. (With Samuel Lepkovsky & Elizabeth A. Murphy.) Jl. Biol. Chem. 107: 429-437- ^ow 1934. The sparing action of fat on vitamin B. \'1I. The effectiveness of various natural fats in sparing vitamin B. (With Samuel Lepkov- sky & Elizabeth A. Murphy.) Jl. Biol. Chem. '07:439-442- Nov., 1934. The sparing action of fat on vitamin G. (With Samuel Lepkovsky & Elizabeth A. Murphy.) Jl. Biol. Chem. 107:443-447. Nov., 1934. The response of the gonads of immature pigeons to \ arious gonadotropic hormones. (With Miriam E. Simpson.) Anat. Recrd. 60:405-421. Nov. 25, 1934. Reduction of the thymus by gonadotropic hormone. (With Miriam E. Simpson.) Anat. Recrd. 60:423-435. Nov. 25, 1934. 1935 The concentration of vitamin G In ailsorp- tion and elution from fullers" earth. (With Samuel Lepko\sky S: \\'illiam I'opper, Jr.) Jl. Biol. Chem. 108:257-265. Jan., 1935. The nutritive value of the lattv acids of lard and some of their esters. (With Samuel Lepkovsky & Roy A. Ouer.) Jl. Biol. Chem. 108:431-438. Feb., 1935. The sparing action of fat on vitamin B. MIL On the loss of vitamin B from the rats tissues. (With Samuel Lepkovsky.) Jl. Biol. Chem. 108:439-455. Feb., 1935. Preparation and properties of vitamin E con- centrates. (With Elizabeth A. Murphy, R. C. Archibald & R. E. Cornish.) Jl. Biol. Chem. 108:515-523. Feb., 1935. Clinical manifestations of dysfunction of the anterior pituitary. Jl. Amer. Med. Assn. 104:464-472. Feb. 9. 1935. Similarity of the urinary gonadotropic hor- mone occurring in some cases of testicular neoplasm with that found in the urine of pregnancy. (With Miriam E. Simpson.) Anat. Recrd. 6i(abstr. supp.): 16-17. Mar. 25- 1935- 'Deficiency' changes in the testicular Leydig cells after hvpophysectomy. (With Miriam E. Simpson &: Richard I. Pencharz.) Anat. Recrd. 6i(abstr. supp.):44. Mar. 25, 1935. Production of superovulation in normal im- mature rats by injection of the principle in menopause urine. (With Miriam E. Simpson.) Proc. Soc. Exper. Biol. K: Med. 32:1046-1047. Apr., 1935. Synergism or augmentation produced by the addition of an hypophvscal synergist to mcno])ause or castration urine. (With Miriam E. Simpson.) Proc. Soc. Exper. Biol. & Med. 32:1047. Apr., 1935. A sensitive biological test lor meixjpausc or castration prolan. (With Miriam E. Simp- son.) Proc. Soc. Exper. Biol, .^v Med. 32: 1048. Apr., 1935. Gonadotropic effects in hypophysectomizcd female rats of implants of pituitaries from castrated males. (Willi .Miriam E. Simpson S: Richard I. Pencharz.) I'roc. Soc. Exper. Biol. & Med. 32:1048-1019. .Apr., 1935. The growth hormone of the anterior pitui- tary. Jl. .\mcr. Med. .\ssn. 101:1232-1237. Apr. 6, 1935. The preparation of crystalline \itamin G. (With Samuel Lepkovsky & William Pop- |)er, Jr.) Jl. Biol. Chem. i()9:liv. Mav. 1935. On the conditions necessary for the contin- uous growth of hypophysectomizcd ani- mals. (With Richard I. Pencharz & Miriam E. Simpson.) Endocrinology 19:509-514. Sept. /Oct., 1935. Clinical manifestations of dvsfunction of the anterior pituitary. (Republication of paper issued earlier in the same year.) In Glandu- lar Physiology and Therapy: a synij)osiitni prepared under the auspices of the Council on Fharniacy and Chemistry of the Amer- ican Medical Association. (Chicago: 1935) 5-28. (Oct.,, 1935. The growth hormone of the anterior piiui- tarv. (Republication of paper issued earlier in the same year.) In Glandular Physiology and Therapy: a symposium prepared un- der I lie auspices of the Council on Phar- macy and Chernistry of the American Medical Association. (Chicago: 1935) 45-59. [Oct.], 1935. The isolation from wheat germ oil of an alcohol, a -tocopherol, having the proper- lies of vitamin E. (With Oliver H. Emer- son c<: Gladys A. Emerson.) Jl. Biol. Chem. ii3:3>9-332- Feb., 1936. Gon:idotropic hormones ... I. AVest. Jl. Surg., Obstel. >^- Gynecol. 11:175-188. Mar., 1936. Gonadotropic hormones... 1 1. West. Jl. Sing.. Obstel. <-v- Gviiecol. 11:199-208. Apr.. 1936. Dietarv prodiulion of specific syiuirome of deficiency in \ii:miin B,.. (With Nellie Hal- lidav.) Proc. Soc. Ex]K-r. Biol. .>v Med. 34: 296-299. .Apr.. i93<^- Gonadotiopic hormones; abstract of Mavo I'oundaiion lecture. Proc. Sialf Mcetgs. Mayo Clin. ii:2i(>-222. Apr. 1, 1936. XXll Bibliography The isolation from cottonseed oil of an alco- hol resembling alpha tocopherol from Avheat-germ oil. (With Oliver H. Emerson & Gladys A. Emerson.) Science (n.s.) 83: 421-422. May 1, 1936. 1 he use of the fractionating column for the separation of fatty acids. (With Samuel Lepkovsky & George V. Feskov.) Jl. Amer. Chem. Soc. 58:978-981. June [5], 1936. Hypophyseal infantilism: treatment with an anterior hypophyseal extract; final report. (With E. Kost Shelton & Lyman A. Cava- naugh.) Amer. Jl. Diseases Childr. 52:100- 113. July, 1936. On the separation and properties of the antagonist, a pituitary substance inhibit- ing ovarian responses to gonadotropic hormones. (With Karl Korpi, Richard I. Pencharz & Miriam E. Simpson.) Univ. Cali- fornia Publns. Anat. 1(8): [iii], 237-253. July 29, 1936. On the separation of the interstitial cell- stimulating, luteinizing and follicle-stimu- lating fractions in the anterior pituitary gonadotropic complex. (With Karl Korpi, Miriam E. Simpson, Richard I. Pencharz & Donald H. Wonder.) Univ. California Publns. Anat. 1(9): [iij, 255-273. Nov. 16, 1936- Fractionation of the gonadotropic hormones in pregnant mare serum by means of am- monium sulfate. (With Karl Korpi, Miriam E. Simpson & Richard I. Pencharz.) Univ. California Publns. Anat. 1(10): ^iij, 275-281. Dec. 21, 1Q36. 1937 Gonadotropic hormone in the blood and urine of early pregnancy; the normal oc- currence of transient extremely high levels. (With Clara L. Kohls & Donald H. Won- der.) Jl. Amer. Med. Assn. 108:287-289. Jan. 23, 1937. On the fractionation of the vitamin Bo com- plex from various source materials. (\Vith Nellie Halliday.) Jl. Biol. Chem. 118:255- 267. Mar., 1937. Effect of interstitial cell stimulating hormone on hypophysectomized male rats. (With Miriam E. Simpson & Richard I. Pencharz.) Anat. Recrd. 67 (supp. 3): 16. Mar. 25, 1937. Marked thyroid hyperplasia without change in respiratory metabolism by the treatment of hypophysectomized rats with thyreo- tropic hormone. (With Evelyn M. Ander- son.)Amer.Jl. Physiol. 1 19:260. June 1, 1937. Dietary production of the syndrome of defi- ciency in vitamin B^. (With Nellie Halli- day.) Jl. Nutrit. 13:657-667. June 10, 1937. On the claim for a new essential dietary fac- tor in mammalian liver. (With Nellie Hal- liday.) Jl. Nutrit. 14:45-54- July 10, 1937- Alleged vitamin E content in royal jelly. (With Gladys A. Emerson &: J. E. Eckert.) Jl. Econ. Entomol. 30:642-646. Aug., 1937. The effect of vitamin E deficiency upon growth. (With Gladys A. Emerson.) Jl. Nutrit. 14:169-178. Aug. 10, 1937. The effect of thyreotropic hormone combined with small amounts of iodine upon the function of the thyroid gland. (With Eve- lyn M. Anderson.) Amer. Jl. Physiol. 120: 597-603. Nov. 1, 1937. The chemistry of vitamin E: tocopherols from various sources. (With Oliver H. Emerson, Gladys A. Emerson & Ali Mohammad.) Jl. Biol. Chem. 122:99-107. Dec, 1937. An anterior pituitary gonadotropic fraction (ICSH) specifically stimulating the inter- stitial tissue of testis and ovary. (With Miriam E. Simpson & Richard I. Pencharz.) Cold Spring Harbor Sympsa. Quantit. Biol. 5:229-240. 1937. 1938 Failure of nicotinic and yeast nucleic acids in "filtrate factor" deficiency in rats. (With Gladys A. Emerson.) Proc. Soc. Exper. Biol. & Med. 38:195-197. Mar., 1938. Growth-stimulating action of alpha toco- pherol. (With Gladys A. Emerson & Oliver H. Emerson.) Proc. Soc. Exper. Biol. & Med. 38:197-198. Mar., 1938. Stimulation of deciduomata around threads on administration of lactogenic and adren- ocorticotropic hormones. (With Miriam E. Simpson & Kaisa Turpeinen.) Anat. Recrd. 7o(supp. 3): 26. Mar. 25, 1938. The purification of the anterior pituitary gro\vth hormone by fractionation with am- monium sulfate. (With Nao Uyei, Quentin R. Bartz & Miriam E. Simpson.) Endo- crinology 22:483-492. Apr., 1938. Atrophy of thymus of the rat resulting from administration of adrenocorticotropic hor- mone. (With H. D. Moon, M. E. Simpson & W. R. Lyons.) Proc. Soc. Exper. Biol. & Med. 38:419-420. Apr., 1938. The calorigenic action of amino acids in the hypojihysectomized animal. (With J. Mur- ray Luck, R. L Pencharz & H. C. Stoner.) Amer. Jl. Physiol. 122:533-541. May 1, 1938- Claim for thyroid subnormality in vitamin E-low rats. (With Ira R. Telford & Gladys A. Emerson.) Proc. Soc. Exper. Biol. & Med. 38:623-624. June, 1938. Bibliography XXI 11 Degeneration of cross striated musculature in vitamin E-low rats. (With Gladys A. Emerson & Ira R. Telford.) Proc. Soc. Exper. Biol. & Med. 38:625-627. June, 1938. Presence of antithyrotropic substance in se- rum of rats injected cfironically with rat pituitary extract. (With Evelyn Anderson.) Proc. Soc. Exper. Biol. & Med. 38:797-798. June, 1938. Steps in the concentration of vitamin B,;. (With Gladys A. Emerson, Ali Mohammad & Oliver H. Emerson.) Jl. Biol. Chem. 124: 377-383- July- 1938- The chemistry of vitairiin E. II. Biological assays of various synthetic compounds. (With Gladys A. Emerson & Oliver H. Emerson.) Science (n.s.) 88:38-39. July 8, 1938- Work performance of hypophysectomized rats treated with anterior pituitary ex- tracts. (With Dwight J. Ingle & H. D. Moon.) Amer. Jl. Physiol. 123:620-624. Sept. 1, 1938. The hypophyseal gro^vth hormone — its sepa- ration from the hormones stimulating the thyroid, gonads, adrenal cortex and mam- mary glands. Resrch. Publns. Assn. Rcsrch. Nerv. & Ment. Disease (Proc, 1936) 17:175- 192. lOct.] 1938. The spirit and task of research. In Addresses: Dedication of Research Building, Abbott Laboratories . . . ([North Chicago, 111.: C1938]) 17-25- 1939 Predominant features of hormonal control of the body by the adeno-hypophysis. (Beaumont Foundation Lectuics, 1938.) St. Paul, Minn.: 1939; 54 p. The occurrence of gamma tocopherol in corn embryo oil. (With Oliver H. Emerson & Gladys A. Emerson.) Science (n.s.) 89:183. Feb. 24, 1939. Endocrine glands: gonads, pituitary, and adrenals. Annu. Rev. Physiol. 1:577-652. [Mar.], 1939. Characterization of gonadotropic hormones of the hypophysis by their sugar and glu- cosamine content. (\Vith H. Fraenkel- Conrat, Miriam E. Simpson & C. H. Li.) Science (n.s.) 89:249-250. Mar. 17, 1939. Relations between the groAvth promoting effects of the anterior pituitary and the thyroid hormone. (With Miriam E. Simp- son & Richard I. Pencharz.) Anat. Recrd. 73(supp. 2): 18. Mar. 25, 1939. Growth-stimulating action of ferric chloride- treated wheat germ oil. (With Gladys A. Emerson.) Proc. Soc. Exper. Biol. S: Med. 41:170-172. May, 1939. Histological changes in skeletal musculature of paralyzed suckling voung of E-low rats. (With Ira R. Telford & Gladys A. Emer- son.) Proc. Soc. Exper. Biol. & Med. 41: 291-295. May, 1939. Rctluced muscle creatine in paralyzed voung E-low rats. (With Ira R. Telford & Gladxs .\. Emerson.) Proc. Soc. Exper. Biol. & Med. 41:315-318. June, 1939. Failure to produce abdominal neoplasms in rats receiving wheat germ oil extracted in various ways. (With Gladys A. Emerson.) Proc. Soc. Exper. Biol. & Med. 41:318-320. June, 1939. Chemical fractionation of the gonadotropic factors present in sheep pituitary. (With H. Jensen, Miriam E. Simpson & Sibylle Tolksdorf.) Endocrinology 25:57-62. July, 1939- Degrees of sterility in female vitamin E-low rats. (\V^ith Gladys A. Emerson.) Amer. Jl. Physiol. 126:1^484. July 1, 1939. Preservation of the seminiferous epithelium and of fertility in male rats by prophylac- tic administration of alpha tocopherol. (With Gladys A. Emerson & Oliver H. Emerson.) Amer. Jl. Physiol. i26:P487- P488. July 1, 1939. The preparation of pituitary growth hor- mone free from lactogenic and thyrotropic hormones. (AVith Donald L. Meamber, Heinz L. Fraenkel-Conrat & Miriam E. Simpson.) Science (n.s.) 90:19-20. July 7, 1939- Preservation of seminiferous epithelium and fertility in male rats on vitamin E-low ra- tions supplemented by a-tocopherol. (\Vith Gladys A. Emerson & Oliver H. Emerson.) Anat. Recrd. 74:257-271. July 25, 1939. Role of vitamin E in the prevention of muscular dxstrophy in guinea pigs reared on synthetic rations. (With Nobuko Shimo- tori & Gladys A. Emerson.) Science (n.s.) 90:89. July 28, 1939. Relation between the growth promoting ef- fects of the pituitary and the thyroid hor- mone. (Willi Miriam E. Sim[)Son X: Richard I. Pencharz.) Endocrinology 25:175-182. Aug., 1939. Action of ketenc on the pitiiiiary lactogenic hormone. (With Choh Hao Li & Miriam E. Simpson.) Science (n.s.) 90:140-141. .-Kug. II, 19.39- The chemistry of vitamin E. XIII. Specificity and relationship between chemical struc- iiire and vitamin E activity. (With O. H. XXIV Bibliography Emerson, G. A. Emerson, Lee Irvin Smith, Herbert E. Ungnade, W. \V. Prichard, F. L. • Austin, H. H. Hoehn, J. W. Opie & S. Wawzonek.) Jl. Organ. Chcm. 4:376-388. Sept., 1939. The effect of cysteine on gonadotropic hor- mones. (With Heinz Fraenkel-Conrat & Miriam E. Simpson.) Jl. Biol. Chem. 130: 243-249- Sept., 1939. Biological studies of the gonadotropic prin- ciples in sheep pituitary substance. (With Miriam E. Simpson, Sibylle Tolksdorf & H. Jensen.) Endocrinology 25:529-546. Oct., 1939- Essentiality of primary amino groups for specific activity of the lactogenic hormone. (With Choh Hao Li, William R. Lyons & Miriam E. Simpson.) Science (n.s.) 90:376- 377. Oct. 20, 1939. Multiple nature of the rat "filtrate factor" — a component of vitamin Bo. (With Ali Mo- hammad, Oliver H. Emerson &: Gladys A. Emerson.) Science (n.s.) 90:377. Oct. 20, 1939- Action of ketene on gonadotropic hormones. (With Ghoh Hao Li & Miriam E. Simpson.) Jl. Biol. Chem. 131:259-266. Nov., 1939. Restoration of fertility in successively older E-low female rats. (With Gladys A. Emer- son.) Jl. Nutrit. 18:501-506. Nov. 10, 1939. Aspects of the function of vitamin E irrespec- tive of its relation to the reproductive sys- tem. Jl. Amer. Dietet. Assn. 15:869-874. Dec, 1939. The vitamin E activity of a-tocoquinone. (With Oliver H. Emerson & Gladys A. Emerson.) Jl. Biol. Chem. 131:409-412. Dec, 1939. Electrophoretic study of pituitary lactogenic hormone. (With Choh Hao Li &: William R. Lyons.) Science (n.s.) 90:622-623. Dec. 29- 1939- 1940 The William Henry Welch lectures. L New light on the biological role of vitamin E. Jl. Mount Sinai Hosp. 6:233-244. Jan./Feb., 1940. Experimental superfecundity with pituitary gonadotrophins. (With Miriam E. Simp- son.) Anat. Recrd. 76(supp. 2):2i. Feb. 24, 1940. Properties of the filtrate factor of the vitamin Bo complex, with evidence for its multiple nature. (With Ali Mohammad, Oliver H. Emerson R: Gladys A. Emerson.) Jl. Biol. Chem. 133:17-28. Mar., 1940. Studies on pituitary lactogenic hormone. L Electrophoretic behavior. (With Choh Hao Li & William R. Lyons.) Jl. Genrl. Physiol. 23:433-43^- M'T- 20, 1940. The effect of thiol compounds on gonado- trophins. (With H. L. Fraenkel-Conrat & Miriam E. Simpson.) Science (n.s.) 91:363- 365. Apr. 12, 1940. The endocrine literature of 1939. (With Bar- bara Cowles.) Endocrinology 26:906-912. May, 1940. Toxicity of high salt intake in adrenalecto- mized rats (With Evelyn Anderson, Michael Joseph & Virgil Herring.) Amer. Jl. Physiol. i29:P30i. May 1, 1940. Growth and reproductive physiology in vita- min B,; deficiency. (With Gladys A. Emer- son.) Amer. Jl. Physiol. i29:P352. May 1, 1940. The prevention of muscular dystrophy in suckling young of E-deficient rats by alpha tocopherol and related substances. (With Gladys A. Emerson & Oliver H. Emerson. j Amer. Jl. Physiol. i29:P354. May 1, 1940. Phosphorus metabolism in the musculature of dystrophic vitamin E-deficient rats. (With Gwei Djen Lu & Gladys A. Emerson.) Amer. Jl. Physiol. i29:P4o8. May 1, 1940. Inactivation of pituitary lactogenic hormone by iodine. (With Choh Hao Li, William R. Lyons & Miriam E. Simpson.) Science (n.s.) 91:530-531. May 31, 1940. Prevention of nutritional muscular dystrophy in suckling E-low rats with alpha-toco- pherol and related substances. (With Gladys A. Emerson.) Proc. Soc. Exper. Biol. & Med. 44:636-639. June, 1940. The prevention of nutritional muscular dys- trophy in guinea pigs with vitamin E. (With Nobuko Shimotori & Gladys A. Emerson.) Jl. Nutrit. 19:547-554. June 10, 1940. Electrophoretic homogeneity of pregnant mare serum gonadolrophin. (With Choh Hao Li & Donald H. Wonder.) Jl. Genrl. Physiol. 23:733-739. July 20, 1940. Experimental superfecundity with pituitary gonadotropins. (W^ith Miriam E. Simpson.) Endocrinology 27:305-308. Aug., 1940. Purification of thyrotropic hormone of the anterior pituitary. (With Jane Fraenkel- Conrat, Heinz Fraenkel-Conrat & Miriam E. Simpson.) Jl. Biol. Chem. 135:199-212. Aug., 1940. Purification of the follicle stimulating hor- mone of the anterior pituitary. (With Heinz Fraenkel-Conrat & Miriam E. Simp- son.) (PPreprint of second following entry.) An. Fac. de med. de Montevideo 25:617- 626. 1940. Bibliography XXV The antagonism to gonadotrophins exerted bv pituitary gonadotrophic: extracts. (With Heinz Fraenkel-Conrat, Miriam E. Simp- son S: Choh Hao Li.) (PPreprint of second following entry.) An. Fac. de med. de Montevideo 25:627-636. 1940. Pinification of the follicle stimulating hor- mone of the anterior pituitary. (With Heinz Fraenkel-Conrat & Miriam E. Simp- son.) (PReprint of second preceding entry.) In Anales de la Facultad de Medicina de Montevideo: numero especial con niolit'o del homeyiaje que la "Sociedad de Biologia de Montevideo" ofrece al Prof. Ludieig Fraenkel, en ocasion del -0° aniversario de su nacimiento. (Monte\ideo: 1940) 159- 168. 1940. The antagonism to gonadotrophins exerted by pituitary gonadotrophic extracts. (With Heinz Fraenkel-Conrat. Miriam E. Simp- son & Choh Hao Li.) (PReprint of second preceding entry.) In Anales de la Facultad de Medicina de Montevideo: numero espe- cial con motivo del homenaje que la "So- ciedad de Biologia de Montevideo" ofrece al Prof. Ludwig Fraenkel, en ocasion del "^0.° aniversario de su nacimiento. (Monte- video: 1940) 169-178. 1940. Further jjuriftcation of the growth hormone of the anterior pituitary. (With Heinz L. Fraenkel-Conrat, Donald L. Meamber ii: Miriam E. Simpson.) Endocrinology 27: 605-613. Oct., 1940. Microscopic lesions without fiuictional im- pairment of striated musculature of suck- ling E-low rats. (With Ira R. Telford & Gladys A. Emerson.) Proc. Soc. Exper. Biol. & Med. 45:135-136. Oct., 1940. Growth-stimulating activity of alpha-toco- pherol. (With Marjorie M. Nelson & Gladvs A. Emerson.) Proc. Soc. Exper. Biol. & Med. 45:157-158. Oct., 1940. Successive generations of vitamin E-lo\v rats. (With Gladys A. Emerson.) Proc. Soc. Exper. Biol. & Med. 45:159-162. Oct., 1940. Purification of the pituitary interstitial cell stimulating hormone. (VV'ith Choh Hao Li &: Miriam E. Simpson.) Science (n.s.) 92: 355-356. Oct. 18. 1940. Interstitial cell siimidating hormone. I. Bio- logical properties. (With Heinz Fraenkel- Conrat, Choh Hao Li &: Miriam E. Simp- son.) Endocrinology 27:793-802. Nov., 1940. Interstitial cell stimulating hormone. II. Method of preparation and some physico- chemical studies. (With Choii Hao Li .^ Miriam E. Simpson.) Endocrinology 27: 803-808. Nov., 1940. Interstitial cell stimidating hormone. Ilf. Methods of estimating the hormonal con- icnl of pituitaries. (With Heinz Fraenkel- Conrat iv: Miriam E. Simpson.) Endocri- nology 27:809-817. Nov., 1940. Purification of follicle-stimulating hormone (FSH) of the anterior pituitarv. (With Heinz L. Fraenkel-Conrat & Miriam E. Simpson.) Proc. Soc. Exper. Biol. & Med. 45:627-()3o. \o\'., 1910. Studies on pituitary lactogenic hormone. II. \ comparison of the electrophoreiic be- ha\ior of the lactogenic hormone as pre- pared from beef and from sheep pituitaries. (With Choh Hao Li 8: William R. Lyons.; Jl. .\mer. Chcm. Soc. 62:2925-2927. Nov. |6|, 1940. Studies on jjituitary lactogenic hormone. I\'. Tyrosine and tryptophane content. (With Choh Hao Li X: William R. Lyons.) Jl. Biol. Chem. 136:709-712. Dec, 1940. 1941 Synergism of estrogens with pituitary gonad- otropins in hypophysectomized rats. (With Miriam E. Simpson. Heinz L. Fraenkel- Conrat )v Choh Hao Li.) Endocrinology 28: 37-41. Jan., 1941. Sliidies on pituitary lactogenic hormone. III. Solubilities of sheep antl beef hormones. (With Choh Hao Li & AVilliam R. Lyons.) Jl. Genrl. Physiol. 24:303-309. Jan. 20, 1941. Effect of pituitary growth hormone on the thymectomized rat. (With William O. Reinhardt .^- Waller Marx.) Proc. Soc. Exper. Biol. & Med. 46:411:415. Mar., 194'- Hormone content of pituitaries of oestrini/ed rats. (With Miriam E. Simpson.) Anal. Recrd. 79 (supp. 2):57. Mar. 25. 1941. Las substancias gonadoliopicas de la antero- hipofisis. Rev. med. de Rosario 31:311-314. Apr., 1941. Influence of lactogenic preparations on pro- duction of traumatic placcntoma in the rat. (With Miriam E. Simpson .<: William R. Lyons.) Proc. Soc. Exper. Biol, .'v: Med. 46:586-590. .\pr., 1941. Growth and graying of rats wilh total iiltrale factor" and with pantothenic acid. (With Gladys A. Emerson.) Proc. Soc. Exper. Biol. is: .Med. 46:655-658. Apr., 1941. Correlation of histological dilTerentialion with beginning of funclion of developing thvioid gland of frog. (Wilh .\ubrey Gorb- man.) Proc. Soc. Exper. Biol. ><: Med. 47- 103-106. Mav, 1941. XXVI Bibliography Studies on pituitary lactogenic hormone. V. Reactions with iodine. (With Choh Hao Li & William R. Lyons.) Jl. Biol. Chem. 139: 43-55. iMay, 1941. Vitamin E activities of some compoinids re- lated to a-tocopherol. (With Max Tishler.) Jl. Biol. Chem. 139:241-245. May, 1941. Anterior pituitary hormones which favor the production of traumatic uterine placento- mata. (With Miriam E. Simpson, William R. Lyons & Kaisa Turpeinen.) Endocri- nology 28:933-945. June, 1941. Studies on the salt-treated adrenalectomized rat. (With Evelyn Anderson & Michael Joseph.) Amer. Jl. Physiol. i33:Pi96-Pi97. June 1, 1941. Phosphorus metabolism of the musculature of E-deficient suckling rats. (With Gwei Djen Lu & Gladys A. Emerson.) Amer. Jl. Physiol. i33:P367-P368. June 1, 1941. O hormonio de crescimento do lobo anterior da hipofise. Resha. clin.-cien. 10:199-201. June 1, 1941. Prophylactic reciuirement for alpha-toco- pherol in male and female rats. (With Gladys A. Emerson.) Jl. Nutrit. 2i(supp.): 15-16. June 10, 1941. La hormona de crecimiento del lobulo an- terior de la hipofisis. Medicina (Buenos Aires) 1:368-374. July, 1941. Effect of the pituitary gro\vth hormone on the epiphyseal disk of the tibia of the rat. (With Robert D. Ray & Hermann Becks.) Amer. Jl. Pathol. 17:509-528. July, 1941. Studies on pituitary lactogenic hormone. VL Molecular Aveight of the pure hormone. (With Choh Hao Li & William R. Lyons.) Jl. Biol. Chem. 140:43-53. Jidy, 1941. Growth hormone of the anterior lobe of the pituitary gland. Jl. Amer. Med. Assn. 117: 287-291. July 26, 1941. Mechanism of action of estrogens on insulin content of the rat's pancreas. (With Heinz L. Fraenkel-Conrat, Virgil V. Herring &: Miriam E. Simpson.) Proc. Soc. Exper. Biol. & Med. 48:333-337. Oct., 1941. Inadequacy for mice of a synthetic diet sup- plemented ^\ith all known vitamin B fac- tors. (\Vith Elizabeth Troescher-Elam.) Proc. Soc. Exper. Biol. & Med. 48:549-555. Dec, 1941. Influence of lactogenic preparations on mam- mary glands and time of vaginal opening in young rats. (With Wra. R. Lyons & M. E. Simpson.) Proc. Soc. Exper. Biol. & Med. 48:634-637. Dec, 1941. Duality of pituitary gonadotrophins. Effects of FSH-ICSH mixture on hypophysecto- mized male and female rats. (With H. Fraenkel-Conrat, C. H. Li & M. E. Simp- son.) Proc. Soc. Exper. Biol. & Med. 48:723- 726. Dec, 1941. The effect of different dose levels of growth hormone on the tibia of young hypophysec- tomized female rats. (With Edwin A. Kibrick, Hermann Becks & Walter Marx.) Growth 5:437-447. Dec, 1941. The early effect of hypophysectomy and of immediate growth hormone therapy on endochondral bone formation. (With Her- mann Becks, Edwin A. Kibrick & Walter Marx.) Growth 5:449-456. Dec, 1941. Time of beginning of function in the thyroid glands of fetal rats. (With Aubrey Gorb- man.) Anat. Recrd. 8i(supp.):95-96. Dec 26, 1941. 1942 The effect of thiol compounds on the activ- ity of lactogenic hormone. (With Heinz Fraenkel-Conrat & Miriam E. Simpson.) Jl. Biol. Chem. 142:107-117. Jan., 1942. Bioassay of the growth hormone of the ante- rior pituitary. (With Walter Marx & Mir- iam E. Simpson.) Endocrinology 30:1-10. Jan., 1942. Effect of purified pituitary preparations on liver Aveights of hypophysectomized rats. (With Heinz L. Fraenkel-Conrat & Miriam E. Simpson.) Amer. Jl. Physiol. 135:398- 403. Jan. 1, 1942. Effect of purified pituitary preparations on the insulin content of the rat's pancreas. (With Heinz L. Fraenkel-Conrat, Virgil V. Herring & Miriam E. Simpson.) Amer. Jl. Physiol. 135:404-410. Jan. 1, 1942. The effect of growth hormone injections on the costochondral junction of the rat rib. (With Robert D. Ray & Her- mann Becks.) Anat. Recrd. 82:6 riij- Jan. 26, 1942. Effect of crystalline estrin implants on the proximal tibia and costochondral junction of young female rats. (With Miriam E. Simpson, Edwin A. Kibrick & Hermann Becks.) Endocrinology 30:286-294. Feb., 1942- Response to growth hormone of hypophysec- tomized rats \vhen restricted to food intake of controls. (With Walter Marx, Miriam E. Simpson & William O. Reinhardt.) Amer. Jl. Physiol. 135:614-618. Feb. 1, 1942. Physico-chemical characteristics of the inter- stitial cell stimulating hormone from sheep pituitary glands. (With Choh H30 Li & Miriam E. Simpson.) Jl. Amer. Chem. Soc. 64:367-369- Feb. [6], 1942. Bibliography xxvu Grow th Iiormone of tlie anterior lobe of the pituitary gland. (Republication of paper with same title issued in previous year.) In Glandular Physiology and Therapy: a symposium prepared uiidcr the auspices of the Council on Pharmacy and Chemistry of the American Medical Association. (Chi- cago: 1942) 19-31. iMar.], 1942. Lack of effect of thyroxin on blood sugar ami glycogen stores of fasted hypophysecto- mized rats. (With Virgil V. Herring & Heinz Fraenkel-Conrat.) Endocrinology 30: 483-484. Mar., 1942. Effect of thyroxin on the insulin content of the rat's pancreas. (With Heinz Fraenkel- Conrat, Virgil V. Herring & M. E. Simp- son.) Endocrinology 30:485-486. Mar., 1942. The bone histology of adult male rats thyro- parathyroidectomized when one month of age. (With Hermann Becks & Edw'in A. Kibrick.) Jl. Exper. Zool. 89:297-303. Mar. 5, 1942. Increased liver arginase on administration of adrenocortical and corticotropic hormones. (With Heinz Fraenkel-Conrat.) Science (n.s.j 95:305-306. Mar. 20, 1942. Lobido-alveolar mammary gro^vth in hy- pophyscctomized rats. (With Wm. R. Lyons R: Miriam E. Simpson.) Anat. Recrd. 82:430. Mar. 25, 1942. Synergism between thyrotropic and growth hormones of pituitary. Body weight in- crease in hypophysectomized rat. (\Vitli Walter Marx R: Miriam E. Simpson.) Proc. Soc. Exper. Biol. S: Med. 19:594-597. Apr., 1942. Urinary gonadotrophins in normal men. (With Aubrey Gorbmau.) Proc. Soc. Exper. Biol. & Med. 49:674-678. Apr., 1942. Biological properties of pituitary interstitial- ccil -stimulating hormone (ICSH). (With Miriam E. Simpson ■<: Choh Hao Li.) En- docrinology 30:969-976. June, 1942. Comparison of methods for standardiza- tion of pituitary intcrslilial-ccll-stimulat- ing hormone (ICSH). fAVith Miriam E. Simpson & Choh Hao Li.) Endocrinology 30:977-984. Jime, 1942. Influence of growth hormone on skeletal tis- sue. Endocrinology 3o:Si()2i. Jiuie, 1942. Effects of crystalline estrin iiuplanis on the tibia of young hypophysectomized female rats. (With Edwin A. Kibrick, Miriam E. Simpson & Herman^n| Becks.) Endocrin- ology 31:93-96. July. "912- Effect of thyroxin and the anterior pituitary growth hormone on endochondral ossifica- tion; species used: the rat. (With Hermann Becks, Robert D. Ray & Miriam E. Simp- son.) Archs. Pathol. 34:334-357. Aug., 1942. EtTects of purified pituitarv prcjiarations on the nonprotein nitrogen constituents of blood. (With Jane Fraenkel-Conrat &: Heinz Fraenkel-Conrat.) Amer. jl. Phvsiol. 137: 200-212. Aug. 1, 1942. CONTENTS I'ACE Kidney-Explantaiion Experiments in Rclalion to Aiiciial Hypertension i Frederick M. Allen The Influence of the Endocrine Organs on Iniesiinal Absorption . . 1 1 T. L. Althausen The Impact of the Introduction of Iron on Medical and Rehgious Thought 25 Walter C. Alvarez The Physiology of the Salt-treated Adrenalectomized Animal .... 33 Evelyn Anderson Contributions to the History of the Bird's Egg Previous to Incubation, by Joanne Evangelista Purkinje 51 Translated by George W. Bartelmez The Localization of Lipids in Cytoplasm 95 R. R. Bensley The Source of Equine Gonadotrophin 105 H. H. Cole and Harold Goss On the Female Testes or Ovaries, by Rcgner De Graaf 121 Translated by George W. Corner The Biological Standardisation of the Vitamins 139 Katharine H. Coward The Prevention of Deafness '49 S. J. Crowe Gene H and Testosterone in the Fowl i57 C. H. Danforth The Influence of Hormones on the Sexual Behavior of Domestic Fowl . 169 David E. Davis and L. V. Domm Pituitary Gonadotrophins i°3 Heinz L. Fraenkel-Conrat, Choh Hao Li, and Miriam F. Simpson Estrogen Assay in the Human *93 S. C. Freed Functional Interrelation of Cerebral Cortex with Basal Ganglia and Cerebelliuii ~*^' John F. Fulton The Solubility of Proteins and Their Separation from Mixtures with Special Reference to Serum - ' 3 Arda Alden Green Gonadotrophic Stimulation of the Ovaries of the Adult Rhesus Monkey 227 Carl G. Hartman The Pathologic, Clinical, and Biochemical Correlation of Tumors of the Testis ~^^ Frank Hinman nxxix:! ^ XXX Contents PAGE History of Hypophysial Diabetes 245 B. A. HoussAY The Symballophone: A Double Stethoscope for the Comparison and Lat- eralization of Sound 257 Wm. J. Kerr On the Significance of the Forgotten Thermodynamic Theorems of Carnot 275 F. O. KOENIG John Banister and the Pulmonary Circulation 285 Saniord V. Larkey and Owsei Temkin Comparison of the Conditions under which Estrogens and Carcinogenic Hydrocarbons are Tumorigenic 293 Alexander LipschOtz Lobulo-alveolar Mammary Growth Induced in Hypophysectomized Rats by Injections of Ovarian and Hypophysial Hormones 315 Wm. R. Lyons Pulmonic Interstitial Emphysema and its Sequelae: An Anatomical In- terpretation 331 Charles C. Macklin and Madge T. Macklin Charles Edward Brown-Sequard 369 Ralph H. Major The Undischarged Ovarian Follicle 379 F. H. A. Marshall Mechanism of the Descent of the Testicle under the Action of Sex Hor- mones 387 Thales Martins A Hemorrhagic State in the Vitamin E-Deficient Fetus of the Rat . . 399 Karl E. Mason Relationships of Sodium and Potassium to Carbohydrate Metabolism . 411 Irvine McQuarrie Harvey's Ideas of Embryonic Nutrition 427 A. W. Meyer Observations on the Pathogenesis of Undulant Fever 437 K. F. Meyer French Medical Education as a Legacy from the Revolution . . . . 461 J. M. D. Olmsted Cytological Differences between Castration and Thyroidectomy Baso- phils in the Rat Hypophysis 471 J. D. Reese, A. A. Koneff, and P. Wainman Studies on the Growth of Lymph Nodes, Thymus, and Spleen in the Rat 487 William O. Reinhardt The Self-Selection of Diets 499 Curt P. Richter Contents xxxi PACE The Relationship of the Anterior Pituitary to the Thyroid and the Adrenal Cortex in the Control of Carbohydrate Metabolism . . . 507 Jane A. Russell Vesalius and Don Carlos: A Historical Footnote 529 John B. deC. M. Saunders Impotence as a Result of Witchcraft 539 Henry E. Sigerist The Coagulation of Blood: Quantitati\e Viewpoints 547 H. P. Smith An Experimental Anatomical Study of Sensory Masking 553 I. Maclaren Thompson The Effect of Progesterone and Lactogenic Hormone ujioa Prolonga- tion of Pregnancy in the Lactating Mouse 561 Kaisa Turpeinen Is Increased Capillary Fragility a Sign of Ascorbic Acid Subnutrition? . 57 1 OsNro Turpeinen The Experimental Production of Pseudohermaphroditism in the Monkey 5^' G. VAN Wagenen and James B. Hamilton The Heart in Myxedema 609 James J. Waring Studies on Growth of Deer Antlers: II. Seasonal changes in the male reproductive tract of the Virginia deer (Odocoileus viygi7iianus bore- alis); with a discussion of the factors controlling the antler-gonad periodicity "29 George B. Wislocki Sex Differentiation in Heterogenous Parabiotic Twins [Ambystoma x Tritunis) ^55 Emil "Witschi and Harriet M. M. McCurdy Ovum, Cycle, and Menstruation 677 Bernhard Zondek KIDNEY-EXPLANTATION EXPERIMENTS IN RELATION TO ARTERIAL HYPERTENSION By FREDERICK M. ALLEN^ M.D. X From tlie DEPARTMENT OF PHYSIOLOGY AND BIOCHEMISTRY NEW YORK MEDICAL COLLEGE, NEW YORK CITY KIDNEY-EXPLANTATION EXPERIMENTS IN RELATION TO ARTERIAL HYPERTENSION AFTER various experiences -with Minkowski's method of explanting the un- /l. cinate process of the pancreas under the skin,'- this method was first ex- tended to the spleen about 1916. The purpose then was to provide a simple means, without laparotomy, of injecting various substances, particvdarly sugar, into the portal circulation. This could be accomplished by injecting into the splenic parenchyma by needle puncture; or by a small skin incision the injec- tion could be made into a branch of the splenic vein. This method was dis- cussed at the time with Dr. H. M. Evans, and though the experiments were broken off because of adverse conditions it seems possible that the method mav still be found useful for some purposes. Explantations of the spleen and kidney were used for otlier purposes in Morristown from 1922 onward. Loesch, Witts and Zimmermann^'^ in their studies of splenic physiology found that the normal volume changes of the organ were hindered by adhesions to the subcutaneous tissue, and, after the return of Dr. Witts to England, Barcroft and Stephens'^ overcame this difficulty by the remarkable device of placing the naked spleen outside the skin and demonstrating that it can survive and function thus indefinitely. Explantations of the kidney were mentioned in several publications" "'^ and also in unpub- lished statements at medical meetings, particularly at the congress on kidney diseases in Minneapolis in 1930. According to the papers published bv Rhoads, Van Slykc, and other writers," '"" they apparently derived suggestions only from the Barcroft and Stephens work," which was by an altogether different method, and knew nothing of the above-mentioned explantations at Morris- town and at the Rockefeller Institute Hospital, of which the "Rhoads method" was a precise copy. Previous reminders'" " of these facts have passed unnoticed. In this renal-vascular research, now resumed after long suppression, kidney explantation is being employed for some of the purposes for which it was originally devised. The present paper will correlate the former work with that now in progress, by giving a synopsis, in the nature of a preliminary communi- cation, of results to be described in a series of forthcoming papers. The topic may be divided into: (1) diet in relation to blood pressure and kidney size; (2) production of acute and chronic hypertension; (3) pathology of the kidneys. 1. Diet in Relation to Blood Pressure and Kidney Size Both femoral and brachial blood pressures are determined by a newly reported auscultatory method" suitable for frequent readings without disturbance. The three dimensions of explanted kidneys can be measured with calipers ^vithin limits of error which are appreciable but still not sufficient to confuse the C ;5 ] 4 Kidney-Explantation and Hypertension actual changes that occur. Table i illustrates typical effects on the blood pressure as observed in individual animals. On the basis of fuller data along these lines in a forthcoming paper^^ it is concluded: (i) The kidneys are subject to transitory functional changes of volume, which according to available evidence are most probably due to hy- peremia. (2) The blood pressure of dogs also fluctuates in connection with feeding. (3) Protein meals may cause no change in kidney volume, or, if the TABLE 1 Examples of Dietary Influence on Blood Pressure and Kidney Dimensions. Dog No. i, Normal; No. 2, with Chronic Hypertension. Dor Diet Before or after feeding Blood pressure Left kidney, cm. Right kidney, No. Femoral Brachial cm. I Bread-chow mixture, f salt extracted | Bread-chow plus 20 gm. f table salt . . . \ Before After Before After 162/88 168/60 156/84 188/92 I 40/80 140/50 138/72 156/78 6.9X4.5X2.4 6.8X4.8X2.5 6.7X4.3X2.2 7.2X4.9X2.6 5.9X3.8X2.1 6.1X3.9X2.3 6.0X3.9X2.2 6.7X4.1X2.8 2 Bread-chow mixture, f salt extracted \ Bread-chow plus 20 gm. f table salt \ Meat, unsalted \ Meat plus 20 gm. table / salt \ Before After Before After Before After Before After 186/108 194/102 190/100 222/90 194/98 188/88 192/98 218/106 I 50/90 145/92 146/84 174/90 148/86 140/80 152/88 168/92 6.1X4.3X2.8 6.3X4.2X2.9 6.2X4.1X2.7 6.9X4.8X3.1 6.2X4.3X2.6 6.4X4.5X2.7 6.3X4.2X2.7 6.8X4.7X3.0 6.2X3.7X2.4 6.1X3.9X2.5 6.0X3.9X2.3 6.7X4.5X2.8 6.3X3.8X2.4 6.5X4.0X2.6 6.1X3.6x2.5 7.0X4.4X2.7 quantity of protein is sufficiently large, may cause a moderate increase, but the blood pressure tends to fall. (4) Salt, water, or especially the two together, regularly produce increases in both blood pressure and kidney volume. (5) The changes in blood pressure and kidney volume are not strictly parallel, inasmuch as: (a) they are not strictly simultaneous; (b) an increase of one may accompany a decrease of the other, as mentioned under protein feeding; (r) there may be discrepancies in degree; for example, with salt the increase in kidney volume is similar in normal and hypertensive dogs but the rise of blood pressure is slight in the former and much more marked in the latter. 2. Acute and Chronic Hypertension In experiments from 1916 onward,"' it proved possible to produce diabetes by brief clampings of the blood supply of a large pancreatic remnant in suc- cessive laparotomies. Such experiments, applied to renal-vascular disease, have a theoretical relation with the arterial spasms which are clinically familiar both in local areas and in the general circulation. The pedicles of explanted kidneys can easily be clamped by inserting the jaws of elastic intestinal forceps subcutaneously, as described by Loesch.^' In Frederick M. Allen the recent resumption of this work a still simpler method has been introduced by the use of either clamps or rubber ligatures applied outside the skin. The results, presented in a paper before the American Society for Experimental Pathology in April, 1941, are divisible into acute and chronic hypertension. Table 2 illustrates the results when the pedicles of explanted kidneys are clamped or ligated by this method. There is an elevation of blood pressure, rising sharply to a maximum, holding a plateau for a variable time, and in the course of several hours declining slowly but not to normal. Whenever the clamp or ligature is removed during this time, the pressure falls rapidly, so that in the course of ten to twenty minutes it is found near or frequently be- TABLE 2 Male Dog, ii kg. Acute Hypertension with Clamping of Pedicles of Explanted Kidneys, for 35 Minutes. Time Blood pressure Femoral Brachial Before clamping 164/66 262/132 212/108 175/78 156/70 148/80 196/120 165/100 145/88 140/92 Before release 5 minutes after release 20 minutes after release 2 hours alter release . . .■ low the original level. There are reasons for regarding this hypertension as nervotis in character, apparently comprising both psychic and reflex compo- nents. The slow^ decline of pressure during prolonged clampings is attributed to nervous paralysis. The sharp fall when the clamps are remo\ed evidently represents cessation of the nervous stimulus. This form of acute hypertension is interesting for several reasons. It may be regarded as the result of an artificial spasm in the renal circulation, although specificity is lacking, as explained below. It is also a hypertension occurring in the absence of any possible substance derived from the kidneys. Further- more, the results are precisely opposite to those obtained under special con- ditions by Taquini" and Prinzmetal.'' They are not necessarily contradictory, because the fact of humoral hypertension, in particular the production of a hypertens've substance in the totally asphyxiated kidney, has been established in carefully controlled experiments. The question whether this substance is identical with that which is produced in partial renal asphyxia of the Gold- blatt type, or whether it can rank as anything more than a post-mortem prod- uct, still remains open. The humoral factor is demonstrated in "pure" experi- ments in which, by anesthesia or otherwise, nervous and other interfering factors are excluded as completely as possible. The experiments with explanted kidneys are important as showing the behavior of the intact animal. In this intact state there is no humoral hypertension following total renal asphyxia. 6 Kidney-Explantation and Hypertension cither because the response is somehow different or because of compensation and adjustment by an overwhelmingly powerful nervous control. If a delayed hypertension follows extremely long ligations of explanted kidneys, such as six or seven hours, there is evidence that this is merely a nervous or psychic phe- nomenon due to the pain of local inflammation. Attempted general theories of hypertension must therefore take account of the complexity of blood pressure regulation. The proof of the specificity of the hypertensive substance obtained from kidney extracts or from the asphyxiated living kidney is that it is not obtain- able from any other organ. One of the proofs of the nervous character of the above-described form of acute hypertension is that it is nonspecific. The re- sults of kidney ligation are fully duplicated by ligations of the testes, legs, tail and other parts. A tourniquet on the tail offers the most convenient routine method, because of absence of the paralysis which results in the legs. With any of these methods there is the same acute hypertension as with the kidneys and also the same rapid fall after removal of the tourniquet. Particularly with ligations of the scrotum for three to four hours and of the tail for seven or eight hours it is possible to demonstrate a secondary hypertension following removal of the tourniquet, due to inflammatory pain, especially in nervous dogs. With this exception, a hypertension following restoration of circulation is as completely lacking with these various organs as it is with the kidney. When the ligations are repeated, whether at intervals of one or many days, the same acute hypertension continues to occur and there is no loss of response with either the kidneys or the tail. But no chronic hypertension results in the case of the tail, as far as observable with frequently repeated ligations for six months. Also, no chronic hypertension has yet been obtained with ligations limited to one kidney, leaving the other intact. Chronic hypertension is ob- tainable, as described years ago by Loesch, by repeated clamping of the pedi- cles of both kidneys, or of one kidney when the other has been removed. The reasons why the original Loesch method is more effective than clamping or ligation outside the skin, also why some dogs develop uremic tendencies while the majority retain good kidney function, and various other questions remain at present undecided because of the long suppression of the work. 3. Pathology of the Kidneys Loesch was not only the first to produce true chronic renal hypertension in dogs by means of the above-described method of intermittent clamping, but also as a pathologist he gave special attention to the accompanying anatomic changes in the kidneys. Figures 1, 2, and 3 are reproduced from microphoto- graphs prepared by Loesch, and his papers sufficiently pointed out the simi- larity of such lesions with the infiltrating and sclerosing processes found in some human cases with hypertension. In the recently resumed research the pathological examination has been undertaken by Dr. W. E. Youland. In view of the intervening developments Frederick M. Allen 7 in the subject, interest may now be foctisecl on the octunence ol hypertension with minimal degrees or even absence of kidney lesions according to the cus- tomary examination of routinely stained sections. Any conclusions must be withheld during a study of possible finer changes, such as the Volhard-Fahr pre-sclerosis, particularly Goormaghtigh's afibrillar juxtaglomerular bodies. Fig. 1. Example of lesions in dogs with chronic hypertension, produced by intermittent clamping of kidney pedicles (Loesch). The Loesch method was not only earlier than that of Goldblaii but may also be held superior to it in many respects as a reproduction of clinical essen- tial hypertension. One of these respects is the absence of the gross atrophy of the kidneys, which in itself sharply distinguishes the Goldblatt conduion from all but rare cases in man. Explantation also furnishes a means of obtani- ing any desired series of biopsy specimens with a minimum of inconvenience or disturbance, and thus of tracing the progress of changes which may be either causative of or secondary to the hypertension. For this purpose an ex- perimental form of hypertension which resembles the clinical essential hyper- tension in the absence of gross renal abnormalities has obvious importance. 8 Kidney-Explantation and Hypertension Summary and Conclusions 1. Explantation experiments in dogs reveal that the kidney is an organ of variable size, with measurable changes dependent upon diet, probably due to changes in blood supply. The enlargement is comparatively slight with protein and more marked with salt and water. Fig. 2. Example of lesions in dogs with chronic hypertension, produced by intermittent clamping of kidney pedicles (Loesch). 2. The blood pressure of dogs, determined by auscultation, is also subject to fluctuations due to diet. The slight rise from salt and water in normal dogs is magnified in those with chronic hypertension. 3. An acute hypertension of nervous character is produced by clamping or ligation of explanted kidneys and also of other organs such as the testes, legs, and tail. The opposite condition which has been described by former writers under special conditions, namely a rise of pressure following restoration of circulation, due to chemical substances flushed out from the asphyxiated kid- neys, never occurs in the intact animal. Frederick M. Allen 9 4. The production of chronic hypertension by intermittent clamping of the pedicles of explanted kidneys, described by Loesch prior to Goldblatt's work, is confirmed. 5. One of the features of superiority in the Loesch iiic iliod is its suiiability not only for the above-described j^lnsiological observations Inii also loi studies of renal pathology, which are now in progress. Fig. 3. Example of lesions in dogs with chronic hypertension, produced by intermittent clamping of kidney pedicles (Loesch). REFERENCES 1. Allen, F. M.: Studies concerning Glycosuria and Diabetes (Cambridge, Mass.: 1913). 2. Allen, F. M.: Jl. Metabol. Resrch. 1:221. 1922. 3. Loesch, }.; Witts, L. J., and Zimmermann, A.: Jl. Metabol. Resrcli. (i:2()7, 1924. 4. Loesch. J., and Witts, L. ].: Jl. Metabol. Resrch. 6:339. 1924- 5. Barcroft, J., and Stephens, J. G.: Jl. Physiol. 64: 1, 1927-28. 6. Allen, F. NL: Jl. Metabol. Resrch. 7/8:217, 1925-26. 7. Loesch, J.: Archs. Pathol. 4:495, 1927- 8. Loesch, J.: Zntrlblt. f. inner. Med. 54:145. '77- '9-53- 9. Rhoads, C. P.: Science (n.s.) 73:417' '93'- 10. Rhoads, C. P.: Amer. Jl. Physiol. 109:324, 1934. 10 Kidney-Explantation and Hypertension ii.Rhoads, C. P.; Van Slyke, D. D.; Hiller, A., and Alving, A. S.: Amer. Jl. Physiol. 110:392, 1934-35- 12. Allen, F. M.: New York State Jl. Med. 38:14, 1938. 13. Allen, F. M.: Trns. Assn. Amer. Physns. 54:80, 1939. 14. Allen, F. M.: Jl. Laborat. & Clin. Med. 27:371, 1941. 15. Allen, F. M., and Cope, O. M.: Jl. Urol. 47:751, 1942. 16. Allen, F. M.: Jl. Metabol. Resrch. 1 :89, 165, 1922. 17. Taquini, A. C: Amer. Heart Jl. 19:513, 1940. 18. Prinzmetal, M.: Jl. Exper. Med. 72:763, 1940. Amer. Heart Jl. 20:525, 1940. Ibid. 21:319, 1941- THE INFLUENCE OF THE ENDOCRINE ORGANS ON INTESTINAL ABSORPTION By T. L. ALTHAUSEN, M.D. From the DIVISION OF MEDICINE UNIVERSITY OF CALIFORNIA MEDICAL SCHOOL SAN FRANCISCO THE INFLUENCE OF THE ENDOCRINE ORGANS ON INTESTINAL ABSORPTION SEVERAL OF THE glaiids ol internal secretion inlluence intestinal absoi ption. In table i are recorded the rates of absorption for dextrose in female rats subjected to removal of the thyroid, the adrenals, the hypophysis, or the ova- ries. These data were obtained by experiments on young rats weighing be- tween 1 50 and 250 gm. which, after they had been fasted for twenty-four hours to clear their intestines, were given by stomach tube known amounts of various TABLE 1 Intestinal Absorption of Dextrose in Female Rats Following Removal of Thyroid, Adrenals, Ovaries, or Hypophysis Experimental conditions iNormal Thyroidectomized. . Adrenalectomized. . Spayed Hypophysectomized . No. of rats Mg. absorbed in I hr. per 100 gm. of wt. 19 lyiiH* 8 9i± 5 15 9 87 ±23 1 23 ±26 17 105^24 * Standard deviation. substances. After a certain period of time, the residue in the digestive tract was determined and the amount absorbed per 100 gm. of weight Avas calculated. After it had been ascertained that a particvdar endocrine organ plays a part in intestinal absorption, experiments were devised in an attempt to determine the mechanism of its action. Our findings are set forth under the headings of the various endocrine organs. The Thyroid Gland. It has been shown by us that intestinal absorption of dextrose not only is decreased by thyroidectomy in rats but also is increased l)y administration of thyroxin to normal animals.' Therefore the effect of the thyroid gland on absorption of various substances was studied in rats rendered hyperthyroid by daily injections of doses of thyroxin that increased the basal metabolic rate to approximately 50 per cent plus. The data in table 2 show that administration of thyroid hormone increases the intestinal absorption of dextrose, galactose, starch, and oleic acid. These substances enjoy preferential absorption through a chenii( ;il mechanism in the intestinal mucosa and are susceptible to phosphorylation. Administration of thyroid hormone does not increase the absorj)ti()n of xylose, alanin, calcium lactate or thiamine chloride-substances which are absorbed by simple diffu- sion and which are incapable of phosphorylation. These experiments indicate that the thvroid gland influences absorption not by increasing the permeability L':-}] 14 Endocrine Factors in Food Absorption of the intestinal mucosa in general but by stimulation of the chemical mecha- nism in the intestinal mucosa which is responsible for preferential absorption. Other experiments ruled out as factors which significantly accelerate absorp- tion: relative starvation, depletion of carbohydrate, a rise in body temperature, an increase in basal metabolism, a rise in the velocity of the blood flow, ac- celeration of gastric emptying, stimulation of intestinal peristalsis, and adjust- ment to an increased food intake. The last factor was shown to be unimportant by an experiment according to which, when food intake of hyperthyroid rats was limited to that of normal rats, the rate of absorption for dextrose remained TABLE 2 Intestinal Absorption of Various Food Elements in Normal AND IN Hyperthyroid Female Rats Time of absorption in hrs. Normal rats Hyperthyroid rats Substance No. Mg. absorbed per 100 gm. wt. No. Mg. absorbed per 100 gm. wt. Dextrose I I I 3 19 8 10 10 I7I±I4* i87±27 I26±24 271 ±44 8 8 10 1 1 284*30 273±l8 196^X4 593 ±125 Galactose Starch Oleic acid Xylose I I 6 2 8 8 10 6 31 ±3 96±6 15. I±2. I 2Izt0.002 4 8 12 6 3 5 ±3 94±io i6.3±i.5 24±0.002 Alanin Calcium lactate Thiamine chloride * Standard deviation. abnormally high (252 ± 10 mg.). A separate experiment with repeated intra- gastric feedings proved that the increased rate of absorption persisted for at least three hours after administration of dextrose was begun. Experiments were carried out on normal and on hyperthyroid rats to test a possible mechanism of action of the thyroid hormone in intestinal absorp- tion based on the theory of Verzar that preferential absorption of sugars and fats is due to phosphorylation in the intestinal mucosa. For the purpose of inhibiting phosphorylation locally in the mucosa, 50 mg. of phlorizin were added to a solution of dextrose, to oleic acid, and to olive oil. The phlorizin reduced the absorption of these substances (table 3) to the same level in normal and in hyperthyroid rats— an indication that the entire mechanism of prefer- ential absorption of these substances was paralyzed. This conclusion was con- firmed by the observation that the reduced rate of absorption for glucose in- duced by phlorizin was equal to that for xylose, which is absorbed by simple diffusion. The possibility that phlorizin may decrease intestinal absorption in general was ruled out by showing that it did not affect the absorption of calcium lactate. In an effort to elucidate further the mechanism of selective intestinal absorp- CQ < <; u -J s to o >■ X h U a. >■ Q <: -5 s o Z s u -3 Q O O > O Z o u b: o ca < -J < 2; K h O 2 N S o J I Oh b< o M O z M -} b. Z r2 '3 a >. X ^ . o E 4^ -H ►- CO c Z 6C 7 o n r", t^ -H M ^ f^ oo (^N V-> a o t "8 oo ON " ri ^ -1^ ^^ -n ^ vc o 0-; >-o cS O Z •±00 <-n r-v o Z o c CO to -So -t- -t- cl « Tt- -t (S ^ ^ 4i 4^ " - \D O r^ r^ ON ■ w C) w U-i c Z ON o o o ^ r^, ^ c3 -a u .- _ _ to o •- P O rt O ^ Q O O U 1 6 Endocrine Factors in Food Absorption tion of sugars, we studied the concentration of the various fractions of acid- soluble phosphate in the intestinal mucosa during absorption of dextrose in normal, in hyperthyroid, and in thyroidectomized rats, and compared it to that observed during the absorption of sodium chloride.* The results indicate that the absorption of glucose is accompanied by a slight increase (15 per cent) in the acid-soluble phosphate esters over that observed during the absorption of sodium chloride. On the other hand, the rate of absorption of dextrose as influenced by the thyroid gland had no statistically significant effect on the concentration of these esters. A positive correlation between the concentration of various phosphates in the mucosa and the rate of absorption would have supported the hypothesis that phosphorylation is concerned with selective absorption of dextrose. A lack of correlation need not militate against this assumption because, if the velocity of formation and the velocity of disap- pearance of the esters are equal, the concentration of the hexose phosphate will be governed by the conditions of the steady state. Accordingly, a differ- ence in the rate of turnover of the acid-soluble phosphates need not be accom- panied by a change in concentration. The question has been raised whether a similarity exists between the mech- anism of absorption of dextrose in the intestinal mucosa and that of resorption of dextrose from the glomerular filtrate in the renal tubide.*'^ In both in- stances absorption involves a selective and active transfer of dextrose across an epithelial barrier at a rate independent of concentration. Also, phlorizin decreases the rate of absorption of dextrose in both organs. Therefore a study of the rate of absorption of dextrose in the renal tubules of three female dogs was made by the method of Shannon and Fisher" before and after the dogs were rendered hyperthyroid. The renal tubules of the first dog resorbed 170 mg. of dextrose (average of four experiments) in the normal state and 250 mg. (average of three experiments) after administration of thyroxin. In the second dog, thyroxin increased tvibular resorption from 200 mg. (average of three ex- periments) to 280 mg. (average of two experiments).^ In the third dog the fig- ures were 214 mg. (average of four experiments) and 335 mg. (average of two experiments), respectively. The possibility that this increase was due to an ele- vation of body temperature was ruled out as far as possible by measurements of the rectal temperature. This experiment supports the conception of an essential similarity between the absorptive mechanism of dextrose in the intestinal mucosa and in the renal tubule. It also strengthens the hypothesis that preferential intestinal absorption of dextrose takes place through phosphorylation because Kalckar* and Colowick, Welch and Cori" demonstrated that extracts of the kidney possess the ability to phosphorylate dextrose. Finally, the data obtained from these experiments offer strong evidence in favor of the specificity of the stimu- lating action of the thyroid hormone on selective absorption of dextrose in the intestine and in the kidneys; they rule out the indirect influence of any physiological mechanism ivhich these tiuo organs do not share. T. L. Althausen 17 The results of our experiments with induced hypcrtliyroicHsni and hypo- thyroidism in rats were confirmed by our work on patients with spontaneous hyperthyroidism and with myxedema. The rate of intestinal absorption in man was determined by administering a solution of galactose by mouth and subsequently measuring at intervals the amount of galactose in the blood. On this basis a new diagnostic test for the activity of the thyroid gland was devised which, in a series of 130 patients with hyperthyroidism and of 97 persons with- out thyroid disease, proved to be as reliable as are determinations of the basal metabolic rate.'" This work also furnished certain therapeutic applications in clinical hyperthyroidism, especially by establishing that glycosuria, postpran- dial hyperglycemia, and high dextrose-tolerance curves in patieius with hypei - thyroidism do not indicate the coexistence of diabetes mellitus unless the fasting blood sugar is also high." This is important because dietary limitation of carbohydrates and administration of insulin are coniraindicated in hyper- thyroidism. Finally, our data point to reduced intestinal absorption as a logical explanation for the low dextrose-tolerance curves in patients with myxedema. Heretofore these usually had been ascribed to greater than normal utilization of dextrose. The Adrenals. Wilbrandt and LengyeP found that adrenalectomy decreases the intestinal absorption of dextrose and of olive oil in rats and that adminis- tration of adrenal cortical hormone restores it to normal. Issekutz, Laszt and Verzar" reported diminished absorption of dextrose and of butter fat from isolated jejunal loops of adrenalectomized cats. Deuel, Hallman, Murray and Samuels" found no impairment in the absorption of dextrose in adrenalecto- mized rats that had received the Rubin-Krick salt mixture. We extended these experiments and found that untreated adrenalectomized rats showed marked impairment of intestinal absorption of dextrose (table 4). After administra- tion of 1 per cent of sodium chloride in drinking water, their intestinal absorp- tion of dextrose was restored to normal; however, the rate of absorption in female rats was not higher than that in males, as is the case in intact animals" (see following section on Gonads). Recently Marrazzi'" in a similar experiment found diminished absorption of dextrose in adrenalectomized rats in spite of the fact that they had received sodium chloride. She ascribed the normal absorption of dextrose in our adrenalectomized animals that had received salt to the fact that we had allowed them access to sucrose during the fust eight hours of the 24-hour fasting period in order to avoid excessive weaken- ing. In control experiments conducted since then with normal rats that hatl access to sucrose in a like manner, we found that the intestinal absorption was not altered. Furthermore, Anderson, Herring and Joseph" demonstrated nor- mal intestinal absorption of dextrose in adrenalectomized rats that had re- ceived salt for periods up to 176 days after adrenalectomy in experiments in which both the adrenalectomized and the control rats had been fasted for twenty hours. In our experience an important condition for successful experi- mentation with animals deprived of an endocrine organ, especially of the M Z O S O <; y P O U < 2: u BS Q < b O Q o; w" Q I— t es O .J u s w PQ < Pi m 5° o CO o "I H < <: M u <" o H u u >J <: u Q < b O CJ u p b. z u K to ++ c 00 rt u ,9 ■a 03 a. tii, J2 • 4^ 4J u < " u. c: < J= d -t- Q Is Z c cT .0 4^ ^ fee r- 0 r^ - u < S J" l-> ^ o i 1 r-^ ; CI ; 00 CO v—*' ^ " * 1-^ • "o' • c:: • • oo r^ T3 D . . : "^ CI •— ' h« e ^-^ m o I- CO G ! '-^ ^ .—^ '-r\ _c ui o c^ f^ d ci "^ -o ^— ' 1 '^ — ^ 1 -T3 rt : : <~* i^ ^ ■+ CJ r : : '^ c< CI n OJ to Ui 0 a ^ 4J O --^ 'o" -o ^ -t- rt . •-0 -^ ^ I^ O ■ '^'-D c-. n ^— -' '■*-i O c u (J '. — - o ^-^ oc I-> . ^^ 'i- *-n r*^ (Lt f) a '~' • ^ ^, ^ f!-; i '. ^- H C-. r-. 3 13 O u ^—^ ,^-^ > ^ ^-'"1 ^ — s ^C ; ^ VC v-C "-o o 00 ' 1 ■ ■ c« ^ -t- ' ^^ oc .9 ; ^c w-i •+ r-. -5 * — ^ "s^^ rt c^ * y i— ^— *s. CTn '^C' -t- o 1 -t- -1- ^ D rt « ^-f :2 s c 4_f ^ ■I-' o C i/; «5 4— ^ w N j-j s ; r- *P o 's : o S 5 w *^ u (J rt oj rt J <; w n! Q < Q M f- «; M ai H I h < M K H O PL, 1^ O O •—I M o X W >- < "^ O 60 e o I— I s o ^ 0 o 'o 2 o; 2 ok o D. -3 2 i^ O I „ oo c o tn s O I-H 1^ 3 nS 2 ^ N o CJ £ TS a, < S i-h" C I— I O o 2 c o O <~o o „ OO ^-v O ON OO M 0\ 2 =k T3 N , o O a. 3 O u o Li CJ CJ Id c CJ u Ul CJ Oh 1-4 O 3 O O J3 c S M 'o > 3J o .r: S D ZaJ Evelyn Anderson 41 potassium in the salt-treated animals with those not so iicated, all ;iiiimals were given 30 cc. Locke's solution containing 6.0 per cent glucose daily ior 24 hours before the urinary collections were made and during the two-day collection period. The Locke's solution was administered by stomach tube, in amounts of 10 cc. three times a day. In table 1 it will be noted that eight days after adrenalectomy the rats which received tap water excreted a greater per- centage of radioactive sodium in 48 hours (60 per cent) than the normal ani- mals (44 per cent); the adrenalectomized rats maintained on 1.0 per cent sodium chloride from the time of operation excreted the same amount of radioactive sodium as the normal rats (45 per cent). Corresponding differences in the excretion of regular sodium were found in the three groups. The adre- nalectomized animals that survived for several months on 1.0 per cent sodium chloride showed a diminishing amount of radioactive sodium excreted per 48 hours. In table 2 it is shown that the excretion of radioactive potassium by the salt-treated adrenalectomized rats was the same as that of the normal animals and it remained normal throughout the four months in which excre- tion studies were done. Thus the alteration in the excretion of sodium and po- tassiimr after adrenalectomy is remedied by the giving of sodium chloride. If the benefit from the salt were nothing more than a replacement of lost sodium chloride, the total quantity of sodium excreted should, if anything, be increased. A recent report by Kottke, Code and Wood'" on urine dilution and concentration tests in adrenalectomized dogs would seem to confirm these findings. They found that the kidney of the adrenalectomized animal main- tained in good condition on a high-sodium, low-potassium diet, when sub- jected to a dilution test, was able to elaborate as dilute, or nearly dilute, a urine as that of an intact animal; however, the concentrating abiliiv of the kidnev of such an animal was less than that of the normal dog. 4. Intestinal Absorption of Sodium, Chloride, and Potassium The rate of absorption of sodium, chloride, and potassium from the intestinal tract is altered after adrenalectomy. Clark"" showed that adrenalectomized rats kept in a healthy condition by salt therapy absorb sodium chloride more slowly than their controls. Moreover, Dennis and Wood" found in adrenal- ectomized dogs maintained on a diet high in sodium chloride and sodiiun bicarbonate and low in potassium a very marked decrease in the rate of ab- sorption of sodium, potassium, and chloride from chronic loops of ileum. The rate of absorption of sodium declined more than that of potassium and often there was an actual reversal in the direction of the net movement of sodiiun. the sodium being excreted into the gut in relatively large amounts although potassium was still being absorbed. These authors suggest that liie finida- mental defect in metabolism resulting from removal of the adrenals may be a decreased ability to perform osmotic work. Stein and Wertheimer" have confirmed these observations in the rat. These observations may possibly throw light on the finding of Anderson, Joseph and Herring^' that adi inalcc- 4-2 Salt Treatment and Adrenalectomy tomized rats maintained for several months on i.o per cent sodium chloride excrete less of an injected tracer dose of radioactive sodium than normal rats given 1.0 per cent sodium chloride for the same period of time. 5. Carbohydrate Metabolism A. Intestinal Absorption of Glucose.— A decrease in the absorption of glu- cose has been observed in adrenalectomized animals. This was first reported by Wilbrandt and LengyeP from Verzar's laboratory. Their absorption studies were carried out on rats between the third and sixth day after adrenalectomy. On the other hand, Deuel, Hallman, Murray and Samuels"" found that the absorption of glucose was normal in adrenalectomized rats which had been kept on the Rubin- Krick salt solution for 1 2 to 20 days postoperatively. The use of salt in the latter experiments accounts for the difference in the results of these two groups of workers, for as Althausen, Anderson and Stockholm"^ have shown, adrenalectomy markedly diminishes intestinal absorption of glucose while the use of 1.0 per cent sodium chloride after adrenalectomy restores the absorption of glucose to normal. Moreover, Anderson, Herring and Joseph^® have found that rats maintained in good health from 10 to 176 days postadrenalectomy are able to absorb glucose almost as well as normal animals. (Tables 3 and 4.) B. Storage of Fed Glucose.— It has been stated by Long, Katzin and Fry^" that adrenalectomized animals that are well fed may show normal levels of glycogen in the liver and muscles. It has not been generally recognized that this capacity of the adrenalectomized animal to store fed glucose is dependent upon an optimal intake of sodium chloride. Anderson, Herring and Joseph"' found that adrenalectomized rats given the stock diet and 1.0 per cent sodium chloride to drink absorbed fed glucose and stored liver glycogen in amounts which approached the normal level; on the other hand, the animals which were on tap water in place of the salt solution had significantly lower levels of liver glycogen even though the absorption of glucose was only slightly im- paired. These data have been summarized in table 3 under groups I and II. These findings lend support to the contention of Britton^" that "in the absence of the adrenal the animal is unable to fix or synthesize glycogen in the liver to any noteworthy degree," but they also show that the giving of salt in op- timal amounts corrects to a large extent this disability. The giving of an excess amount of salt diminishes the ability of the adrenalectomized rat to store fed glucose. This is shown in group III of table 3. The rats of this group were given 4 cc. of 5.0 per cent sodium chloride four times a day. The adrenalec- tomized rats so treated were in a state of health comparable to adrenalecto- mized rats on a low salt intake. The adrenalectomized rat on an optimal intake of salt retains the capacity to absorb fed glucose and to store it as liver glycogen for several months after operation (Anderson, Herring and Joseph"**). The data on this are summarized < > < U 0. o o Oh Q K f- •z (d W O < en a u N S o u u oi Q < Q Id H < M «: H I H o < a: o u < Q >- X o ta > 1- 4^ 4^ 4^ 4^ . r-5 r^ r^ Lo r o tj 1 Qj 60 ^ CTs O - SO CI 1 ■t B ^ -. ^ _, hJ -< — C4 r-: oo vo vo I^ CO 1 0 u ' >-"u 4 4 d r^; oo O T(- >-o r-^ f^ cl r-- ^s. -H -H + H ^ 4^ u do >-r\ oo <^, O •1- wo r-, Vy-i -t- ^O Cl c-j vo w-» »-o r^ -^ i-T *^ v-C CI r~~ o^ wn ON 6CJ! vv-> vo -i so O rf VO 1 \0 <-o r-) M » « P- " M ^ -ti ^ + 41 o Ir u ■1-' ^ 4> r^j oo 1-1 . c y 13 -o z 4-1 13 t- tr 1-. 1) u H ■J ^ 13 ^i-13 ^ J^ "3 V 13 Z ?^ ■-5Z o 13 u -o c — CJ SJ N 1^ -a c 4-J £ o £ ^ = 2 ^ > o en CJ c t. '-'^ U C I. u 5 "o U C t. 1— < ^,13 o tiU o'~r c o 3 ~; O Q u N s o o M <: 2; w o: O < o M E- < H I h < O VI W b! O h U < b! Q >< K O pq < ^ 7, 8 E ►J tio-y "be V ~ be « c 3 U ^ ti II ■Sao S-o' d T 4j C o -^ E E-? OO ct O -o OO d OO so "-o r~- o so "J-i ^ so -H -H ii -H n OO en CO •-0 lO O OO d d ^ 4^ -t ON 1-1 d 4^ a d o - w-i CO >-o O CI CI d - r^ >-<-i CO M 4^ 4^ OO Vj^ 4!" 4^ ct 0 CI 4^ so CO -H CO d w. ON ON CO 00 ON ON CO 00 ON ON On CO 00 r-~ CI d O so -t CO o u O +-" .a £ ii E S-^ °, In >= 0.« 1- 1/^ -^ « « "5 D C/l Q < Z O t-( a o X W u o o ci H O -a 1 I ^ =3 CIS S3 h- i ^g -a • ^3 n • OJ T3 U". "rt 1 o S^ ^3 ^ >> 13 u ^o 1 O vo^ ^3 S3 Ul J, Ml C 00 c s O I ^3 ft3 Sn3 c CO >. >. 1-. « Q 'C 3 50 ? 00 " Co" c 3 cS 3 §^3 rt Ui OJ > < t- ^ ^S o?3 §;S Ln CI .— ^ g^3 d ,— s 1 CI O ^'^ O ^ _,_ m •^ ^— V -s /""^ S;3 C^J '^s 1 CI ^ &(:s O >>D * 4J — 6C«^- l^ -^ 2 CO 1— 1 2|^ >. c^ d Vy ^ 7 t^ 7* D > n rt cl' 1 1 l-H 1 t-i 1 ^ - i; -o d OC oo »— 1 CO < S O' d ^ ^ ^ O • r^ • 4; CO ii O • 0 6 ■ 6 • C ^ *c5 3 l>^ Q T3 O • d OC o • d r~- • lU N 05 CO 05 4-^ *-> +-> rt a oi )_ u v> ■g -d -a 13 o (U (U in whicli the specimen had been immersed, the hillock described above was drawn to the dry central boss at the bottom of the dish, it collapsed and almost distintegrated so that its central poms was dilated in all directions to almost double its own diameter. When I examined it under a lens I was filled with no little wonder as a most exquisite vesicle appeared, free in part, in part adhering to the margin of the porus. Now it ^\as not difficult to separate it and to isolate it com- plete and intact (figs. 5 to g). So it is that the poms which appears to penetrate the hillock of the cicatricula is occupied by a very small perfectly clear vesicle which, immersed in a substratum filled with globules, presents only two free surfaces, one toward the external vitelline membrane, the other to- ward the interior of the yolk where at the summit of the hillock it is surrounded by a small crater. The relations are nevertheless such that its external surface clings a bit to the vitelline membrane so that it is easily torn when they are separated, whereas on the inner surface a single layer of mutually adherent globules is continued over it from the hillock; so it is that the appearance of an opening [poms] is simulated. When you have succeeded in uncovering this vesicle compleiclv you will find that it has the shape of a somewhat flattened sphere cn- l^Sl ^ eloped by a most delicate membrane and containing a perfectly clear watery flidd. Sometimes a fringe of the white material of the hillock persisted in sticking to its lateral periphery but usually a little ma- nipulation completely freed it of this. Traces of adherent vitelline membrane were sometimes to be seen (fig. 8). There was still some question as to whether the vesicle of the hillock grew fast on all sides to the material at the marginal zone and whether the edges of the poms were closed by special membranes. Many observations leave no doubt but that the vesicle is discrete and distinct. If one may bring the imagination to the aid of the adroitness of senses and hands I would ventiue the opinion that a membrane [con- sisting of] globules extends from the margin of the poms of the hillock [cumulus] to the periphery of the vesicle and that the membrane is continued all over both the outer and inner hemispherical surfaces of the vesicle, the cuticle of which remains intact. So it is that the cicatricula of the ovarian egg has a distinctive and characteristic structure in the form of a somewhat compressed sphe- roidal vesicle which consists of the most delicate of membranelles in- closing a characteristic watery fluid, perchance generative in nature (and so I would call it vesicida germinativa). It is intrenched in a white mammiform hillock composed of globides, perforated at the center [by the poms]. The cimndus is surrounded by a flat whitish 64 Translation from Purkinje zonule which is then continued into the layer of globules, referred to above, which invests the inner surface of the vitelline membrane. It may be added that the cicatricula is so closely applied to the inner surface of the vitelline membrane that the two always remain in con- tact when separated from the yolk. The distinction between the cicatricula of ovarian ovules, oviducal eggs and laid eggs. My overhasty imagination [imaginatio praecox] was already divin- ing that this vesicle was tenanted by the female germ from which the chick in its turn would develop. The next thing then was to investi- gate with the utmost care the cicatricula of a recently laid egg before there had been any incubation whatever, to see what transformation our vesicle had undergone in it. An entirely new state of affairs was found here since the cicatricula [^ 4 ] of the newly laid egg does not stick to the vitelline membrane although that of the ovarian egg separates from it with difficulty. On the other hand, the cicatricula of the latter is easily separated from the under- lying yolk while that of the laid egg is closely attached to the yolk. Furthermore, the zona of the cicatricula of the ovarian egg is still thin and in closer relation to the colliculus where the formative sub- stance is gathered, as it were; however in the cicatricula of the laid as well as of the oviducal egg, everything is spread out diffusely to all sides, the colliculus has already been dissolved, the semi-transparent blastoderm is everywhere uniform in thickness, nor is there any trace of the vesicle to be found. The cicatricula of the fully formed egg takes the form of a double circle, of which the outer is fused to the yolk, the inner, which is continuous with it, is separated from the yolk by a shallow little fossa (perchance Malpighi's "colliquamentum")."' This circular little fossa persists in the yolk and presents at its center a whitish knob (^nucleus of Pander) covered with a viscid semi-trans- parent substance through which white farinaceous granules are scat- tered. Similar granules also besprinkle the inner surface except at the center (fig. 1 1) which rests upon [Pander's] nucleus but does not stick to it. This central part of the blastoderm is so semi-translucent that the darker region of the fossa has a leaden color while the nucleus shimmers white (figs. 9-11)- It would seem therefore that the vesicle of the ovarian egg had burst in the laid egg and had been converted into a fluid [colliquamentum]. It was still [necessary] to investigate the cicatricula, while the egg was in the uterus, and while the yolk still remained in the oviduct. In the former I have found the same conditions as in the laid egg. But in George W. Bartelmez 5r the yolk within the oviduct there was no trace of the genninal vesicle, although at the beginning, while it was still clinging to the infundib- ulum, the remains of the colliculus were present, easily separable from the yolk. So it ai)pears that the vesicle is burst or dissolved by contractions of the oviduct when the semi-fluid yolk is taken up by the infundibulum and that its fluid is so mixed with the substance of the colliculus that from it that fluid [colliquamenliim] with white granules is produced; from the rest of the colliculus, the nucleus [of Pander] is formed.* §3 On the development of the germinal vesicle (Keimbldschen). [^ 5 ] The next job was to investigate this same vesicle in the cicatriculae of still smaller eggs, an easy matter, especially in cases where the yolk material has not yet attained the usual density and flows like milk. Then when the ovum is cut open under water, the membrane is promptly freed of yolk. On its inner surface a diaphanous vesicle pro- trudes slightly, surrounded by a narrow halo of white globular mate- rial, which in the mature ovum constitutes the cumulus. At this stage the vitelline membrane is of the softest texture and is covered with a rather thick layer of globules. The vesicle does not increase in size from the beginning to maturity in proportion to the egg as a whole. In the smallest it is scarcely less than half the size it attains in mature eggs, so that it then may fill almost all the space destined for the yolk. If you were to look back over the course of development, you would say that the vesicle is the first structure in the ovule to be stimulated to germinate but at a slow rate which later and up to maturity is far surpassed by the yolk and its envelopes, so that from the mathematical point of view there are two series of equal length beginning with equal volumes, one of them increases more slowly, the other more rapidly, so that while the ratio is at first practically one of equality they eventually differ from one another many thousand times in mag- nitude. In ovules that have reached the diameter of one line and less, a circular spot corresponding to the vesicle is to be seen from tiie ex- terior under a lens of medium power, shadowed forth [from the yolk- laden ooplasm] by its transparency [umbrosa ob pelluciditatem] (figs. 12-15). It is easily detected in ovules of 4 to 6 lines where it stands out immediately when the membranes are cut open under water and the vitelline membrane is removed. Here I must call attention to * [This note was added in the 1830 edition.] It now seems more probable to me that the vesicle forms the basis of the dark central part of the blastoderm referred to above (fig. 11) and that its hemisphere is expanded into a double membrane [that is, the two-layered blastoderm]. 66 Translation from Purkinje the passage in Tiedemann* where he says that the cicatricula is com- parable to the Graafian ovules in mammals, because it is the first struc- ture to develop in the ovule and that it has the form of a compressed vesicle or white spot. It is certainly not difficult to remove this vesicle of ours in small eggs where it is not yet surrounded by white globular material; in mature ova this is a more difficult undertaking. The fol- lowing method, which I hit upon by chance at first, was of the greatest help to me later. After you have very carefully separated the cumulus with its porus from the vitelline membrane under water, using a probe or curved needle, allow it to settle to the bottom of a dish which must have a convex bottom; then as the water is pipetted off, the cumulus will spread and the vesicle will stand revealed, if not at first, [^6] at least after several trials. But you must be exceedingly careful lest you suck up the specimen with the water and destroy it. Indeed, the consistency of the vesicle is so delicate that in the smaller eggs it may burst like a water bubble at the slightest touch. §4 The vesicle of the cicatricula as a normal female organ in birds. In reflecting on the significance of our vesicle, the first thought that arose was that it might be a sign of mating and that it would not be found in the ovules of hens which had not been kept with the cock."' However, it was also found in isolated hens, all of wdiose eggs were examined to make sure, and found to be infertile. In some it seemed to me more delicate and much more difficult to isolate; in others abso- lutely no distinction was found. §5 Concerning the development of the yolk and its central latebra. With regard to the development of the yolk, I find only a few things to be noted. At the beginning, clear miliary vesicles can be observed projecting out in great numbers from the ovarian surface like a skin eruption. At this time the yolk is a clear fluid material not easily differentiated from the vesicle of the cicatricula. By the time the ovule has attained the diameter of one line, the whitish yolk becomes turbid and the vesicle appears transparent through the external membranes. \Vhen it has reached the size of a pea, the yolk has taken on the color of thick yellowish milk. Then oily droplets can be recognized, though * Zoologie, Vol. Ill, p. loo. [Friedrich Tiedemann (1781-1861). His text was pub- lished in 3 volumes between 1808 and 1814. Vol. Ill has the subtitle Anatomie und Naturgeschichte der Vogel.] George W. Bartelmez f,- rather infrequently, since a sero-albuminous fluid iMcdominaics. With increasing size the yolk becomes more yellow, uiiiil in the lully ma- ture egg it has attained almost a golden color. The yolk in the mature ovule, and while it is still in the beginning of the oviduct, is found to be much more viscid and cohesive than it was previously and than it afterwards becomes while the albumen is being laid down; perhaps it takes up some of the fluid of the latter/"' Even at the time when an ovule begins to grow yellow, there may be recognized at its center a whiter and more fluid material such as you will And rather closelv [ 7 ] resembles the cicatricula in mature ovules. This white yolk material is best seen for the first time in laid eggs.""' It is this no doubt which in its time gave rise to the problem of Bellini,* for it is found not only in boiled eggs but also in the very freshest and in every ovule from the time that it becomes clearly visible. Thus, when you have cut off parts of the yolk under water with a pair of Cowper's scissors, begin- ning at the cicatricula, the fresh surface exhibits three concentric zones of different colors;"" the outer of these is rather pale, the middle a deeper yellow, the inner again paler surrounding the latebra of more fluid white material. This fluid is situated at the center and appears to send out a delicate canal toward the middle of the cicatriculaf (figs. i6-i8). At all events this white center with concentric strata surrounding * [This footnote was added in the 1830 edition, but the "problem of Belhni" is dis- cussed in the 1st edition.] See Laur. Belhni: Opuscula aliquot ad Archibald. Pit- carnium, Lugd. Bat., 1714. p. 14: and further: Comment. Bonouiens. \o\. II, Pt. I, p. 85; Pt. II, p. 369 and following; Comm. Pauli Bapt. Balbi de lielliniauo prob- lemate; also Miscell. soc. Taurinerts. Vol. I, p. 3 and following in Commentario Joannis Franc. Cigna, who was the first to refute Bellini in the matter of the cicatricula of boiled eggs.<"> [The first of these references is to an "opuscuhim" addressed to Archibald Pit- cairn in 1693, namely De motu cordis intra & extra utcrum. Bellini's great discovery is set fortli on p. 112 of the Opera Omnia, Part II (cd. of 1708). Balbi's contribution occurs in De Bononiensi Scienliariim et .irtinum Institulo atqiie Academia commentarii. 1731. Cigna's refutation of Bellini is in Miscellanea pliilosophico-mathematica Socie- tatis Privalae Taiirinensis, Vol. I: "De Belliniano jjroblcmati sen de ovorum elixa- torum cicatricula."] t [Footnote added in 1830.] Later investigations have raised doubts in my mind concerning this structure. Von Baer, to be sure, in his work {Ueber die Entwicke- lungsgeschichte der Thiere, Erster Theil, Konigsbcrg, 1828) has applied our ob- servations on the central latebra of the yolk to his theory of halos,^^> but whether he has confirmed it by his own observations does not appear in any published work. Unless the arrangement of the chalazae and their firm attachment to the volk by the intervening membrane of Dutrochet suffices to explain the constant position of the cicatricula at the upper surface, I should .say that the central fluid with its canaliculus extending to the cicatricula acts like a plumb line so that the yolk, tending of its own accord to rise upward within the mass of albumen is prevented from further rolling so that the cicatricula always preserves the same position uppermost. At any rate it can be justly assumed that the watery fluid [of the latebra] has a higher specific gravity than the semiolcaginous substance of the yolk. 68 Translation from Purkinje it, is so similar in form to the cicatricula that it is no wonder that for some time an absurd opinion confused the one with the other, espe- cially since in the hard-boiled egg the albuminous cicatricula could hardly be distinguished from the albumen to which it adheres. The white substance contained in the central latebra of the yolk, as seen under the microscope, consists of rather large white globules suspended in an albumen. It has almost the same consistency and structure as that which makes up the fluid part of the cicatricula, so it would seem that there is some developmental relationship be- tween the two regions. The rest of the latebra is filled with a rather fluid lymph. It is more convenient to look for this structure of the yolk in boiled eggs, so long as you are sure that it was not produced [; 8 ] in the first place by the cooking. At times between the first and middle strata [of yolk] I have found a spherical lamella of coagulated albu- men, or here and there albuminous projections from the center, espe- cially in eggs that had already been incubated for some hours. In the boiled yolk it appears that the central latebra often presents an irregu- lar figure and at the same time the concentric strata have correspond- ing irregularities. In the boiled yolks the central latebra contains a milky substance somewhat salty in flavor so that from this it is clear that it differs chemically from the rest of the yolk. In many eggs I have found the inner layer of yolk rather oily, translucent and deep yellow, but I have not yet discovered on what conditions this de- pends. What further changes the yolk may undergo during incuba- tion, how the halos around the cicatricula are produced in it, how the albumen is gradually mixed with it, must remain for later investiga- tions. For the sake of those who enjoy microscopic observation, I may add that the boiled yolk, which readily disintegrates into a powder, shows masses of the most beautiful corpuscles, simulating crystals in form, and indeed those removed from different parts of the yolk show different patterns. §6 Concerning the vitelline membrane. The vitelline membrane is uniformly delicate and exceedingly transparent and shows no organic structure under the microscope, no matter how it is cut up or teased. It is a perfectly intact closed vesicle, pierced by no opening. In the ovarian ovule, to be sure, you can scarcely differentiate it in early stages from the layers of globules which are laid down by the yolk peripherally; at that time also the vesicle of the cicatricula is so joined to it that it is not possible to separate one from the other without injury. Even in older ova, the membrane of the germinal vesicle sticks so fast to the vitelline mem- brane that very often in separating them you tear the latter, leaving George W. Bartelmez Bq a circular fragment (fig. 8*). Thus the vesicle seems, as ii were, to nestle on the vitelline membrane itself and to develop with it.*"' So long as the ovule remains in the ovary, its external membranes are so thick as compared to the exceedingly delicate vitelline mem- brane that only the most persistent efforts suffice to separate them com- L 9 H pletely from the vitelline membrane, without injuring it. It would be well worth while to inquire into the relation between this membrane and the inner surface of the external vascular membrane in the ovar- ian ovule, how it comes about that the blood provides for the secre- tion of the materials of the yolk. It is at least certain that the whole inner surface of the vitelline membrane is, at this time, covered with clear globules uniform in size, resembling blood corpuscles in form and easily differentiated from oily yolk globules. The vitelline membrane while it is still in the infundibulum ap- pears wrinkled and quite pliable so that the yolk, which is more viscid and of a firmer consistency at this time, can take on the extremely elongated shape necessary for it to pass the rather narrow canal of the infundibidum to the oviduct Avith its looser walls, without any tearing of the vitelline membrane. §7 Concerning the membranes of the calyx. Now while the ovarian egg is approaching maturity, the petiole is elongated to almost lA ^ thumb's length, so that the most mature ovum hangs down beyond the rest toward the uropygium and is pressed into the wide open mouth of the infundibulum by the under- lying intestines aided by the abdominal muscles.""' Then the stigma of the ovum becomes more and more pale and while it is not thinned out, it nevertheless becomes softer so that at last it is very easily torn and releases the ovum. At this time, the external vascular membrane contains a loose mesh of broad thin-walled veins, beginning on both sides of the stigma with the branches disposed like very fine combs. It appears to be a resorbing mechanism by which the tissue of the stigma is gradually and imperceptibly dissolved and ovulation is prepared for.""' The stigma does not first appear in the latest stage of the ovide but it is already present as a fine, diill, whitish line in ovules of Vis the full size which are still semi-transparent. The internal lamina of the follicular membrane which adjoins the vitelline membrane is not of a veil-like texture, as some would have it, and as it does indeed ap- pear superficially, but it is very smooth and transparent and witliin it minute corpuscles are scattered, which are perhaps glandular <.. vascular tufts fairly regularly distributed. '"' 70 Translation from Purkinje §8 [ 1 <0 Concerning the movements of oviduct and infundibulum and their muscular mechanism. When you open a recently killed hen which has an already fully developed egg in the uterus and you remove the intestines so that the egg-bearing organs are fully exposed, you will observe the whole oviduct and the uterus writhing in continuous peristalsis; especially if you immerse it in warm water, the ruffled fimbriae which crown the free margin of the infundibulum are curled by the most beautiful alternating contractions and expansions. But if you seek for the true seat of the movements by stimulating the parts of the oviduct you will find it with little effort in the mesometrium. No one who looks carefully will deny that this is provided with numerous unmistakable muscle fibers.* Perhaps it will not be superfluous to devote one or two lines to the description of this apparatus. In the laying hen the mesometrium is quite different from the mesentery which merely conveys vessels to the intestine. It is a true muscle or rather a muscular membrane, which to be sure also distrib- utes vessels to the oviduct. Two parts of the mesentery are to be recognized, an inferior and a superior. The inferior is inserted on the lower surface of the uterus if we may so term the lowermost part of the oviduct where the shell is formed and which plays the major role in the laying of the egg. Here a cross-shaped, rather compact plexus of muscle fibers (fig. 19, c) spreads out about the uterus from both sides; at the posterior part of the uterus, where the vagina is inserted into it, a rather slack muscular sac is formed, which at laying encompasses the vagina dilated by the egg and helps in its delivery. At the other end, however, it is dilated into a muscular membrane which is reticulated with fibers that spread like a fan; its periphery begins at the insertion of the oviduct into the uterus and ends at the insertion of the posterior angle of the infundibulum into the uterus, thus folding back on itself. Now the other, namely the anterior, angle of the infundibulum is gathered into a rather compact elastic ligament, perhaps wholly mus- cular, which is attached for the most part at the root of the penulti- mate rib of the left side but elsewhere it is attached less conspicuously [^11] by many delicate projections about the pulmonary tubes [poros aeri- feros pulmonum] (fig. 19). In the midst of the ligament the upper part of the wall of the abdominal air sac [sacci aeriferi abdominalis] is attached on all sides (fig. 19, h). * [Footnote added in 1830.] See Ge. Spangenberg: Disquhitio [inaiiguralis ana- tomica] circa partes genitales foemineas avium (Gottingen: 1813), p. 50. J I George W. Bartelmez h i From this ligament* the sheet of the superior nicsometrium takes its origin. Together with the air sac it arises from the peritoneum at the left side of the vertebral column and passes on to the dorsal wall ■ of the uterus. From here muscle fibers descend to the superior part of the oviduct on the surface of which they are so thinned out that they are hardly visible, and, forming the exceedingly delicate muscu- lar tunic of the oviduct, they extend to the inferior mesomctrium. The internal glandular tunic of the oviduct is thinned out to the utmost in the infundibulum, ending at its fimbriated border; the extension of the mesometrial muscle on the other hand is more developed so that this border seems to be composed entirely of muscular tissue.f As a matter of fact, the muscle fibers of the fimbria [limbiis] form a very delicate and complex net; toward its margin the meshes become smaller and more numerous, ending as very fine ones in the somewhat crenated margin (fig. 20). The principal direction of the transverse fibers is perpendicvdar to the margin of the infundibulum. In the mid-sagittal line of the infundibulum the muscle fibers run together fiom both sides of the fimbriae so that they produce a transversely striated suture; bttt w'here the mouth of the infundibulum gapes open they are continued through its walls to the oviduct. In addition, other fibrils parallel to the margins of the fimbriae are interwoven at right angles to the fibers mentioned above. These fibrils take their origin from the margins of the infundibulum, namely from the anterior ligament and from the uterus, and extend on both sides through- out its length. When these fibers contract, they bring about the utmost wrinkling of the fimbriae. The remarkable mobiHty of the infundibulum mentioned above is ascribed to these fibers. The rest of the mesometrial muscle fibers are prolonged transversely to the oviduct, communicating with one another and mutually intertwining by means of elongate folds and branches so that transparent areas [12] remain. These areas consist solely of the layers of the serous mem- brane and are also interwoven with very delicate ner\e fibers that run transversely and with secondary oviducal vessels that intersect at sharp angles; the principal vessels, however, which are closest to the origin of the mesometrium, run parallel to the oviduct. The -^vidth of the mesometrium is not everywhere the same. For, as the OA-iduct bends up or do^vn in three flexures, the mesometrium also becomes shorter or looser. Tlie muscle fibers of the mesometrium. just as they reach the oviduct, diverge from each other and pass about it, surrounding it with a very dclioite muscular membrane, in ^\•hich you will hunt in vain for the longitudinal fibers of which authors * Perhaps the taenia of Spangcnbeig (as cited), p. 55. t Spangenbeig (as cited), p. 55, seems to maiiiiain llic tonliary. 72 ' Translation from Purkinje speak* except in the lowest part of the oviduct where, passing over into the uterus, it is constricted for the extent of about 2 thumbs' breadth. -f- Here perfectly distinct longitudinal muscle fibers extend from the uterus and they promptly vanish in the region where the oviduct widens out again. The uterus itself is covered by a double muscular layer. In the external layer the fibers run in a longitudinal direction from the end of the oviduct to the beginning of the vagina; in the internal layer they encircle the meinbrane which secretes the shell for the egg. Longitudinal fibers also predominate in the vagina; the circular fibers are rather g^^thered together into its sphincter. The vagina and infundibulum are remarkably distensible where the canals are almost wholly muscular, so that by gradually dilating them you can easily introduce several fingers. Here distensibility has been asso- ciated with contractility, since they are adapted to the moving for- ward of bodies and, at the same time, holding them without their escaping although they are close to the external openings. §9 Concerning the taking up of the ovarian egg by the infundibulum of the oviduct.;]; The further history of the mature ovule may now be considered [^ 13 ] in the hope that it may throw light upon a field so buried in darkness. At the time when the yolk has attained the requisite size, the peduncle of the calyx is elongated to such an extent that, moving back between viscera and ovary, it touches the opening of the infundibulum. If, after removing all the viscera, you look at the ovule hanging from the ovary, you will see its less rounded surfaces turned to right and left, the stigma however looking downward. But if you now replace the viscera, they press the ovule to the mouth of the oviduct in such a way that the flattened parts are turned up and down, whereas the stigma is actually kissed by the mouth of the infundibulum. The muscular structure and well known mobility of the infundibulum * [Footnote added in 1830.] These longitudinal fibers are simply folds of the in- ternal albumen-secreting membrane of the oviduct, which shimmer through the serous membrane and muscle as very delicate lines. Once these membranes have been dra^vn aside, it appears to be easy to smooth out the folds of the secretory membrane by opening up the folds with a spatula. f [Footnote added, 1830.] The isthmus of the oviduct I would term it. This part, as I have said, is adapted for the secretion of the shell membrane. While the ovum, now surrounded with albumen, is entering the beginning of the isthmus, the first part [of the albumen] is compressed by the resistance of the walls as it enters and thus the sharp end of the egg is moulded; as its journey continues, the passage is already dilated so that the albumen is rounded off at the blunter end; in this constriction we should perhaps look for the [origin of the] narrower strand of albumen which they call the ligament of Tredern. X See Spangenberg (as cited), p. 59 and following. \ George W. Bartelmez >nc, and its fimbriae easily prove that they arc not inert at this lime when they embrace the ovule by licking it on all sides. As I understand the situation, it is as follows: The mouili ol the infundibulum is dilated by its longitudinal muscles and it takes up the ovum in its entirety, the fimbriae on the other hand are wrinkled by the fibrillae which run down the sagi tally directed infundibular margins and are constricted about the stalk of the ovule. Thus the calyx of the ovule is, as it were, softened both by the motions of the infundibulum and its secretions. When the blood flow is impeded by the constriction and pressure, and all nutrition is stopped, the calyx is stretched even to the point of rupture of the stigma; then the yolk is rolled out to be taken up later by the infundibulum. At that lime the characteristic transverse fibrils of the infundibulum, which form the suture described above, as well as the fibers at the beginning of the superior and inferior mesometrium, contract alternately toward the interior of the canal and force the yolk completely out of the open calyx."*' The yolk is now propelled along the infundibulum and ar- rives at the intermediate part of the oviduct which secretes albumen. The infundibulum then opens again and expels the empty calyx. Such an idea of the advance of the ovule by the oviduct seems fairly sinely determined by the anatomical structure of the organs, although it rarely ever happens that the process is actually witnessed in the living animal. It is unlikely that these movements are accomplished by a swelling up of the infundibulum, since it does not present that vascular spongy structure peculiar to erectile organs, nor is it neces- sary to resort to this explanation if we consider its muscular apparatus. [14] I have found that in recently killed hens the position of the in- fundibulum is very variable with respect to the oviduct. In one case the infundibulum hangs down perpendicularly, in another it falls over to the left, in another to the right side; it is not necessarily turned to- ward or away from the oviduct,* but it is suspended near the latter, at- tached by its extremities to the left penultimate rib and to the uterus. § 10 The superabundance of secretion, and the involution of the reproductive organs. In many hens which carry a fully developed egg in the uterus, I find the surface of the most mature [ovarian] ovum already macerated as it were, wherefore I should venture to infer that it has experienced some softening from the activity of the fimbriae. At this time you ^vill * See Al. Monro: Versuch einer Abhandlung iiber Vergleichende Anatomic, ans dem Englischem (Gottingen: 1790), p. 80. [Alexander Monro primus: An Essay on Comparative Anatomy (London: 1744; 3d ed. edited by his son. Ale.\. Monro II, 1782).] 74 Translation from Purkinje usually find in addition a yellowish fluid in the abdominal cavity as if mixed with a kind of albumen solution, and you will not infre- quently find the same fluid poured out between the layers of the calyx. I do not know what interpretation to place on it; it is perhaps to be derived from an excessive effort of the blood to secrete yolk, like the post-partum lactiferous abscesses in women. Never, to be sure, have I found a mature ovum so ruptured that the yolk was diffused between the layers of the calyx and the spherical form of the ovule distorted into an irregular one. If, however, laying hens are under-nourished, the formation of eggs ceases and those ovules which were approaching maturity undergo a peculiar transformation; the inner membrane of the calyx is ruptured, the yolk flows out into the spaces of the outer membrane, the ovule decreases in size, the yolk becomes whitish and is resorbed; at that time even the smaller ovules lose their spherical form and hang, flabby and wrinkled, devoid of yolk. It is remarkable how quickly the muscle fibers vanish in the meso- metrium of fowls which have stopped laying, so that even after a few days you can discover only traces of them, milky in color and hardly visible. But these matters require much detailed discussion. Concerning the disappearance of the germinal vesicle. After the yolk has been received by the oviduct, it is invested during its progress with chalazae, the chalazal membrane, albumen, the [ 15 ] double shell membrane, and the shell. You will best understand the internal structure of these if you follow the changes the egg undergoes as it passes through the oviducal canal. If, in the first place, you investigate the cicatricula after the yolk has been taken up by the infundibulum, you will nowhere find the vesicle which has been described above in the cicatricula of the ovule. Instead of the porus an internal circle of the blastoderm is now to be seen, the cumvdus seems to have been changed into a white center, and between the two a small circular area will be noted, besprinkled with white gianvdes. If a mechanical explanation of this transforma- tion be adequate, I would say that the yolk, in bursting from the calyx and in being taken up by the infundibulum, is subjected to such disturbance by the contractions of the latter that the exceedingly delicate vesicle is ruptured. However, an account based on an opinion ivill not fill up gaps in observation }^^ Repeated observations must be made with the utmost acumen, unless perchance this naturally difficult matter has withdrawn itself among mysteries not to be approached. George W. Bartelmez 75 § 12 0£ the formation of albumen and chala/.ae. I turn now to the later stages in the formation of the egg. At the upper end of the oviduct, where the longitudinal folds of ilic mucous membrane begin, a first thin layer of albumen appears enswaihing the vitelline membrane; at opposite ends where the oviduct is con- stricted above and below [the ovum], soft, clear, albuminous nodules are laid down, from which on both sides a strand of albumen is con- tinued, surrounded by the folds of the internal lamina of the oviduct; these strands are the rudiments of the chalazae (fig. 21). Up to this time there is no hint of the white twisted strands which occupy the center of the fully developed chalazae, nor is the vitelline membrane as yet marked by radial folds through the twisting of the chalazae. Now, as the yolk is moved along by the peristalsis of the oviduct, layer upon layer of albumen secreted by the walls is added in the form of a spiral band. Meanwhile, the first lamina of albumen begins to thicken against the yolk and is converted into a rather firm envelope, adher- ing very closely to the vitelline membrane. From this envelope the internal whitish strands of the chalazae extend. The yolk with the surrounding albumen is movable about its [longest] axis. The chalazae attached to the membrane [of chalaziferous albumen] are produced [] 16 ] by twisting, as a thread is spun from a distaff and twisted upon itself again and again. Accordingly you will also find spirally wound layers on the very surface of the albumen which is subsecjuently laid down. It ^\'ill be most obvious that this is not a figment of the imagination if yoti will take out an egg already fully surrounded by albumen from near the lowest part of the oviduct and study it after about an hour's immersion in perfectly fresh cold spring water. The albumen which was at first clear and transparent is gradually made turbid and opales- cent by the water and shows strands on its surface which run spirally from left to right proceeding from the blunt to the more pointed end. If you tease them away with volsella forceps, they come off in lamellae all runnina: in the same direction until the whole albumen is un- wound down to the chalazae (fig. 22). Now Avhen you have come almost to the chalazal membrane which covers the yolk, the last layers are very soft, so that you almost think the space empty; the rest of tiie albumen passes to the chalazae and hangs like a hood from them. From these facts it may be inferred that Fabricius ab Aquapendente proposed a perfecdy correct interpretation of the formation of albu- men and chalazae, and no one should believe that they arise from the vitelline membrane and grow like plants.*""' 76 Translation from Purkinje § 13 Of the chalaziferous membrane of Dutrochet. At first the chalaziferous membrane of Dutrochet* is assuredly al- bvimen. In its progress through the oviduct it becomes solid and finally takes on almost the same character as the vitelline membrane. The internal whitish strand of the chalazae is the continuation of the chalaziferous membrane and is prolonged from it to the interior of the chalazae like a tube twisted on itself and closed by torsion, just as if you were to invest a sphere with a membraneous sac open at both ends and close it by twisting at both poles. The vitelline membrane itself remains unchanged and intact as it was in the ovarian ovule. There is no communication through it with the chalazal canal by which albumen might be conducted to the [] 17 ] interior of the yolk.'"" The canal which Leveillef saw can be readily demonstrated if you will but cut across a chalaza under water. The demonstration is successful only in chalazae which are doubly twisted on themselves so that the mutually opposed turns necessarily form an inner canal like the modiolus of the cochlea. This nevertheless does not communicate with the yolk in any way (fig. 22). Although the vitelline membrane does not appear porous under the microscope, it seems permeable everywhere to nutrient fluids, just as is the case in the ovarian ovule where it takes blood from the vascular [thecal] membrane, transfers it to the interior and converts it into yolk mate- rial (fig. 22*). We see the same thing happen at the beginning of the development of the chick when the albumen above the cicatricula disappears and enters the interior, producing a crater. We readily infer, therefore, that a similar thing probably happens to albumen over the whole surface of the vitelline membrane; there is no need to conjure up suctorial ducts. It is not difficult to separate the chala- ziferous membrane from the underlying vitelline membrane in the place where the chalazae are inserted, even where that part of it which is twisted into a strand opens up in the [chalaziferous] membrane. To do this successfully you should take the chalaza near its vitelline root with a pair of volsella forceps and at the same time cut from the yolk hanging below it a piece of the vitelline membrane, wash it, and then, under a medium-power lens tear the two membranes [vitelline * Joiir7ial de Physique \de Chimie, d'Histoire Naturale et des Arts], Vol. 88, p. 170 [to 178, 1819:— Henri Dutrochet: "Histoire de I'oeuf des oiseaux avant la ponte." Translation in Deutsch. Arch. f. Physiol. 6:379-385, 1820.]. f Reils Archiv [/»r die Physiologic], Vol. 4, p. 418 [1800; this is an abstract of J. B. F. Leveille: "Dissertations physiologique sur la formation des foetus dans les mammifdres et dans les oiseaux" /. de physiq., de chim., d'histoir. natiir. et d. arts, Ann. 7 de la Republique (1799), p. 386]. Geome W Bartelmez ° / / and chalaziferous] apart under water, using two paiis of forceps. There are eggs in which it is obvious to the naked eye that the chalazif- erous membrane is loosely attached to the vitelline [membrane] only at the insertion of the chalazac (fig. 23). There are others in which the central strand does not end at the vitelline membrane but at an interpolated globule of albuminous substance covered by the chalazif- erous membrane.'-* § 14 Of the whitish zone on the surface of the yolk. In a great many eggs (fig. 24) one or several white strands extend from one or the other central chalazal strand through the chalazif- erous membrane on the surface of the yolk, running in no regular w^ay, now on one side, now on the other, or encircling the whole yolk, as if the membrane had been thickened or folded there. These con- [ 18 ^ stitute what Vicq d'Azyr* calls the zona albicans; they are very differ- ently disposed in different eggs and do not, as some claim, have a constant form and relation to the yolk. They are nothing but unmis- takable remains of the chalaziferous membrane. These strands usually run out from the smaller chalaza at the blunter pole, which being sometimes twisted obliquely passes over into them. There are fre- quently many of them running over a part or the whole of the surface of the yolk, avoiding in general the cicatricula; but sometimes they pass over the middle of it. If you tease off the chalaziferous membrane, it is sometimes possible to demonstrate folds; but usually they are not folds but thickened matter of the membrane itself, giving a silvery sheen like a tendon. At times strands of this kind are to be found only at the blunter pole of the egg and there is no sign of a chalaza. j- Then the albinnen, too, is found to be much diminished at the blunter pole of the egg, almost all of it being gathered at the sharper end. If you were to look into the origin of this condition in the oviduct, you would say it is due to a too rapid secretion of albumen, namely that the part of the yolk which precedes in the passage down the oviduct stimulates its membrane to a copious secretion and quickly exhausts it so that when the upper end of the yolk moves into the same region little is left to be secreted. These conditions in isolated accidental cases also explain why, under normal conditions, there is a smaller chalaza at the blunter pole of the egg as well as a lesser amount of albumen. * Oeuvres, Vol. 4, p. 392. [Fel. Vicq d'Azyr: Fragmem sur I'Anatomie et la Physio- logie cle I'Oetif, etc. (Paris: 1805).] f See Leveiile (as cited), p. 416. 78 Translation from Purkinje § 15 Concerning the so-called third albumen and the zone of the chalazae. In investigating the question of a third albumen, beware lest you delude yourself as others* have done and as I, too, did for a long time. It is true, of course, that the albumen about the central whitish strand of the chalaza is thicker and almost approaches gelatin, and if you wish to call it the third albumen to distinguish it more precisely, I have no objection, although it does go over gradually at the extremi- ties of the chalazae into the rest of the albumen so that it cannot be [[19] delimited in any way. The mound, however, which the cited author figures about the chalaza and which is continued in the form of a girdle to the other end of the egg, where it surrounds the chalaza, and also what he calls the "ambient chalaza" and the "third albumen," all this is not real, but an oblique view of the outer surface of the second albumen, closely applied to the yolk and chalazae after the rest of the albumen has settled to the bottom of the dish by its own weight and diffused. In my opinion, at least, it seems to present an optical illusion because of a peculiar surface. Those who used to talk about a normal girdle of the chalaza surrounding the cicatricula and embracing the yolk sphere [dividing it] at a ratio of 80:100 seem to have suffered from a similar illusion. §16 A membrana propria of the albumen is denied. Writers are also wont to talk about a membrana propria of the albumen, but I greatly doubt whether anyone has seen it unless he has first produced it artificially, since fresh water forms it by coagu- lation wherever the water has touched the albumen. I have succeeded best in making a membrane of this sort apparent by the following procedure: You keep a fresh egg after removal of the shell immersed in cold water for several hours. The albumen then develops a rather solid whitish surface. You may now suck off the water surrounding the albumen and insert a slender tube into it as far as the root of the chalaza and inflate the albumen by blowing gently. It gradually swells up into a bubble almost 6 times the volume of the egg. If the operation goes off well, it retains its shape and remains intact. You may then again gently pour on water so that the ovum, now floating freely, may be expanded on the under side where the bottom of the dish pre- ' * Comes ab Tredern [Louis Sebastien Marie de Lczerec de Tredern]: Diss[ertatio Inauguralis Medica] sistens Ovi Avium Historiae \et Incubationis] Prodr\omium. (Jenae: 1808)], Figs. 2 and 3. [German translation by L. Stieda, Wiesbaden: 1901.] George W. Bartelmez ^o viously compressed it. You then see the yolk with chala/ae ahiiosi free, except where at their ends they pass on both sides into the albumen, all included in a large expanded egg-shaped sac, fairly firm and float- ing on the surface of the water. If you cut into the surface gently, it is possible to strip off from it a fairly thick whitish membrane. A new membrane of the same kind is formed as often as you pour on water or spirits of wine. § 17 C 20 ] Concerning the fluid albumen. So long as the egg remains in the oviduct, the first or fluid albumen is not present. Neither is it to be distinguished in eggs within the uterus, even when they are already covered with a shell. The second albumen at that time sticks to the shell membrane and no Ugamentum albuminis is to be recognized at the more pointed end. The first albu- men does not diffuse when the newly laid egg is opened under water but, remaining ovoid in form and more transparent, it surrounds the second albumen, although it is still at that time much more fluid. The effect of atmospheric air seems to be of chief importance in the formation of this albumen, as in the formation of a blood clot. Per- haps even the air pressure draws the more fluid part from the inter- stices of the lamellae of the second albumen toward the periphery. It is certain that before the more fluid part has been separated from it, the second albumen is much more readily penetrated by fresh water in ^vhich it is placed and the lamellar structure is brought out. The fluid albumen possesses a very gieat plasticity, as appears w^hen you open the egg under fresh water and after it has diffused it promptly coagulates into exceedingly delicate films and fibers. If you wish some pleasant amusement, break an egg into a flat dish without adding any water; the fluid albumen which has spread over the bottom of the dish is easily distinguished from the thicker albu- men surrounding the egg. You may then suck up part of the former in a slender tube and drop it into fresh water or a solution of galls diluted with water, or you may pour it so that it settles slowly to the bottom. Then the most exquisite saccules, tubes, membranes, cells, and fibers are produced, which you can hardly differentiate from true organic structures, and the process goes on to the point where natin e seems to build up structures of her own. § 18 Concerning the origin of the chalazae and their mechanical use. While the yolk is being invested with albumen in the oviduct, it is moved along gradually until it reaches the isthmus. Its progress is not like that of the intestinal contents propelled by muscular walls. The 8o Translation from Purkinje muscular fibers of the mesometria play the most important role as they draw the oviduct hither and thither, bellying, twisting, and con- [^21^ stricting it. At this time the part of the mucous membrane which en- velops the yolk secretes albumen which takes the form of membranes and promptly sticks to the yolk; on the other hand the part before and behind the ovum is constricted and moulds the secreted albu- men into a column, which is twisted upon itself as it is moved along spirally and is gathered up at both poles of the egg; as a result of this tQrsion it is in general drawn toward the root by which it is attached to the yolk, and there it merges into the rest of the albumen (fig. 24). Since the chalazae are inserted on the ends of the axis of the yolk, their purpose seems to be chiefly to keep the axis rotating always in a single plane. So it comes about that the cicatricula is always the uppermost part of the yolk and nearest the heat of the incubating hen. If you rapidly twirl an egg in a mechanical device, the chalazae do not acquire more convolutions but the twisted ligamentum albumenis is torn from the shell. § 19 Concerning the formation of the shell membrane. The shell membrane begins to be laid down on the external sur- face of the albumen in the lowermost part of the oviduct, which lies between its isthmus and the uterus. First the sharper end enters and is invested with [shell] membrane so that at the blunter end the al- bumen may often be seen still bare. The isthmus region is so stretched in this period of activity that the folds of the lining membrane are completely wiped out and there is very great tension in this distinctly narrower part of the oviduct. Perhaps the greater irritation of the membrane elicits a denser secretion which produces the membrane. The brief interruption of the mucous folds at the level of the isthmus constriction should also be noted. At this point the transition between the albuminous and membrane secretion is made, and here perhaps the conditions are provided for the formation of the ligamentum alhuminis at the sharper end of the egg. The shell membrane is laid down double; the inner is composed of straight fibers of microscopic dimensions, the outer presents no distinctive structure* (fig. 25). * In the egg of Coluber Natrix this internal layer consists of the most beautiful wavy fibers. See G. R. u. L. Ch. Treviranus: Verm[ischte] Schriften [Anatomischen und Physiologischen Inhalts (Gottingen: 1816-21).] Vol. 1, p. 142. [The particular essay referred to is by Gottfried Treviranus (1776-1837): "Ueber die organische Elemente des thierischen Korpers." The description and figure refer simply to "Eierschale" so that the observations on Coluber were Purkinje's own.] George W. Bartelmez 8 1 Concerning the position of the egg in the C 22 ] uterus and the formation of the shell. I have always found the position of the egg, newly arrived in the uterus, such that the sharper end is pointed toward the vagina, the blunter end toward the interior. In the fully formed egg, however, when an effort has already been made to lay, I have sometimes found the blunter, sometimes the sharper, end applied to the opening of tlic vagina."'' Perchance in the labor of laying the egg it is often turned until it assumes a comfortable position. When the shell is beginning to form, you will find its membrane at first besprinkled with the most minute calcareous flakes all about the same size and polygonal in form (figs. 26 and 27). Then they keep accumulating and fusing together, but so as to have inconspicuous interstices Avhich serve for transpiration. If now you will examine an oviduct which has been opened up from its infundibulum to its entrance into the cloaca, you ^vill find its various parts differentiated for the following functions: The mouth of the infundibulum receives the yolk as it is extruded from the calyx; the following part of the oviduct, which comprises about 14 of its total length and which is provided with a rather scant glandular apparatus, secretes the membrane of Dutrochet and the innermost strands of the chalazae; farther along, the internal membrane of the canal is covered by a mucous-producing parenchyma considerably thickened. This comprises almost two of the foin~ parts of the whole or more, and provides the great investment of albumen for the yolk. Thereupon the oviduct is narrowed at the isthmus and in the folloAv- ing portion as far as the uterus, [comprising] almost I3 of the whole, the shell membranes cover over the albumen. The uterus, in its tin n, serves to secrete the shell; the vagina extrudes the egg through its own external opening and through the cloaca. The purposes of these organs are readily visible and there is no need of calling in hypotheses to build up a "counterfeit presentment" of nature. 82 Translation from Purkinje TRANSLATOR'S NOTES (i) This is a cordial and generous reference to von Baer's Epistola^ announcing the discov- ery of the mammalian ovum. It included a complete confirmation and extension of Purkinje's observations" (see translator's note 5). Anyone who has studied the various types of yolk granules in the bird's egg can readily understand how natural it was for von Baer in 1827 to homologize the "granules" of the mammalian "cumulus oophorus" with the white yolk granules of Purkinje's "cumulus" (see Sarton^"). Von Baer saw the germinal vesicle in young mammalian follicles but failed to see it in his "ovulum." He concluded that the entire Graafian "vesicle" is homologous with the oocyte of the hen and that the "ovulum" is the transformed germinal vesicle. In 1833 Coste" reported at a session of the Paris Academie des Sciences that he had identified the germinal vesicle in the rabbit's egg and this was noted a month later in Froriep's Notizen. Purkinje refeired to these observations in his 1834 article^^ "Ei." Since his own efforts to find the germinal vesicle in mammals had been in vain, like Coste's earlier ones, he was not pre- pared to accept the finding at the time. In the following year, however, Valentin and Bern- hardt in Purkinje's laboratory confirmed Coste and this was published in Bernhardt's thesis"" (1834). In the 1834 memoire,^^ Coste reported the disappearance of the germinal vesicle before ovulation in the rabbit, although in the previous year he thought he had recognized it in an oviducal egg. Purkinje's suggestion that the disappearance of the germinal vesicle at the time of ovulation be investigated did not bear fruit for fifty years. At the time the whole situa- tion was fraught with insuperable difficulties, and Purkinje's fear that "this naturally difficult matter might be an unapproachable mystery" was justified (see Meyer,^" chap. VIII). Von Baer thought he had seen the germinal vesicle, as such, extruded from the ripe frog's egg. He was probably observing polar body formation. After his convincing evidence that the germinal vesicle is no longer present just before ovulation in hen and frog he would have been confused indeed if he had known that it was still intact in the tubal eggs of the dog. This is one of the few vertebrates in which the first maturation division does not begin until after ovulation (Van der Stricht,^^' ^^ Evans and Cole'^). The understanding of maturation had to wait upon the discovery and interpretation of mitosis and on the unraveling of the histo- genesis of the spermatozoon and they in turn depended on the development of staining methods. (2) See Bartelmez," p. 328. (3) Purkinje does not say here which of the "lines" he means. There were some eight "lines" in use in various European countries at the time, all in the neighborhood of 2.2 mm. In his article "Ei"^" (1834) he gives a series of measurements of hens' oocytes in "Vienna" lines and his collaborator Valentin"^ in 1835 employed the same unit. The Vienna line corresponds to 2.195 mm. It ^vould be interesting to know whether the line developed from Aristotle's smallest unit of measurement, a millet seed. A hundred of the dry seeds which I ha^e measured averaged 2.85 mm. in length and 1.82 mm. in width. (4) Purkinje mentions this term casually as if it were familiar to his readers. When he in- troduces a ne^v name he is careful to call attention to the fact, so it must be assumed that tliis one had been used before. I have, however, failed to find it anywhere except in the two editions of the present work. Valentin in his textbook^" (18351 implies that the observation of "a clear agranular and perfectly transparent spot at the center of the cicatricula had been made by Fabricius ab Aquapendente, Harvey, and others and had been readily repeated by many naturalists." There is nothing in Fabricius, or Harvey, or any eighteenth-century author I have read, to indicate that the "poms" had been seen in the hen's egg. There are several figures in the literature before 1825 which show what can readily be interpreted as the germinal vesicle of other eggs. The best of these are Poll's figures in his sumptuous work'*" on "testaceous" molluscs (1791; see pi. 16, fig. 18, showing a group of young oocytes from the George W. Bartelmez §,. sJ ovary of Cardium). The description of fig. i,, pi. 3, reads: "F.odem microscopio amplificala et piincto veluti lucido centrali imiguila, rcpraesanlur." It seems to be indicated by E Geoffroy Saint-Hilaire=>^ in a drawing of a hen's ovary as a clear spot at the center of the cicatricula in two large oocytes. There is no reference to it in the legend or text. In 1822 Home.^'s describing drawings made by F. Bauer, referred to "an aperture in the inner mem- brane" of the ovarian oocyte of the hen, which, he says, "has not been before taken notice of." Prevost and Dumas in their paper^» on the chick (1827) cited Geoffroy Saintriiiaire and Home and said (p. 421), "Les jaunes de I'ovaire ont une cicatricule trt^s apparent ... lille est marquee de deux cercles concentriques et d'un point plus transparent qui en occupc Ic centre. Ce dernier semble produit par une ouverture de la membrane du jaune." Obviously they knew nothing of Purkinje's "Gratulationsbrief" of i82r,. It is interesting to note that this interpretation differs from the original ideas of Provost. as indicated by Dumas in his 1825 article^ in which he wrote concerning ovarian hens' eggs: "Au centre de la cicatricule, on observe un point de couleur jaune, et d'apr^s les observations recentes de notre excellent ami le docteur Prdvost de Geneve, celui-ci est du a une petite vesicle transparente, entierement semblable a celle qui se rencontre dans la come des mam- miferes des les premises jours de I'accouplement." Furthermore, Prevost, in a paper" on the development of Unto (p. 450), had described a "clearer disc" in the ovarian oocvies. Prevost's original idea of a vesicle or disc seems to have been replaced in 1827 by the notion of an aperture, possibly because Dumas was more impressed by Home's remarks than by his good friend's conclusions. It would appear then that Prevost and Dumas missed the discovery of the germinal vesicle by as narrow a margin as they missed recognizing the mammalian ovum (see Corneri^). Whoever may have seen the "poms" first, there is no question but tiiat Purkinje's isolation of the vesicle was the first evidence that it is a discrete structure within the egg. (5) In the 1825 edition (p. 3) this passage reads: "Poms hie in cicatricula ovomm enixorum, et qui in iitero atque in oviductu latent nusquam invenitur." In 1830 "enixorum" (that is, "laid") is changed to "auctomm" (that is, "grown" eggs). Von Baer had stated in 1827^ '^i^t he had in a single instance found a ripe ovarian egg of the hen in which the germinal vesicle had disappeared (p. 28): "Semcl enim in ovo gallinaceo maturo, in ovario quidem relenlo sed ad ejectionem parafo, vesiculam non inveni." In 1828 he had demonstrated the dog's egg to Purkinje, A. Retzius, J. Miiller, and E. Weber (see Corner,^" p. 86) and this matter of the disappearance of the germinal vesicle may weW have been discussed at that time. In any case, Purkinje stressed the matter in the 1830 "Praefatio" and changed the text. (6) The students of spermatozoa and of blood appear to have been the first to note the untoward effects of water on living cells. Charles Bonnet" in 1779 states that he had observed in 1773 "Si on fait toucher sur les animalcules du spermc humain une goutte d'eau de pluie oil meme d'eau distille, ils perdent a I'instant le mouvement, mais ils le conscrvent dans la salive chaude ou froide." William Hewson," in 1777, described the change of form and the plasmolysis of red blood corpuscles on the addition of water to blood. Purkinje found, as soon as he had his compound microscope (1832), "class manche Gegenstiinde in ihrer Form und Gestalt durch Wasser veriindert werden" (article-" on "Ei," 1834, p. 114). He recom- mended the use of the thin (outer) albumen of the hen's egg as a mounting medium. In the first edition of his Handbuch der Physiologic^ (1834) Johannes Miiller referred casually to the effects of water on red blood corpuscles and recommended diluting the blood with weak salt or sugar solutions. J. Henle^^ in 1841 says regarding the study of fresh tissues, "Nicht ohne bittere Erfahrungen hat man gelernt, dass die Wahl der Flussigkeit, deren man sich ... bedient, nicht gleichgiiltig ist" (p. 143). He advises the use of aqueous and vitreous humors. blood serum, and dilute solutions of sugar, cooking salt, and other neutral salts. This seems to have led giadually to the general use of normal salt solutions. The earliest study of the exact concentrations of salts necessary to prevent cell injury that I have found is that of de Vries*' (1871) on the cells of the red beet. 84 Translation from Purkinje (7) This is the fluid of the subgerminal cavity which had been of great importance since Aristotle in the discussions of the interchange between the "moist" albumen and the "earthy" yolk (see Fabricius in translation of Adelmann,^ p. 176 ss.). (8) The fact that a hen may lay fertile eggs for weeks after isolation from the cock was known to Aristotle and was responsible for much of the mysticism surrounding the process of fertilization. William Harvey believed that the "aura seminaUs" fertilized the clutch of eggs in the ovary. This idea is not compatible with Aristotle's statement that if a hen is mated and subsequently trodden by a second cock "while the eggs are still yellow," that is, still in the ovary, the chicks will resemble the second cock. It is quite possible for part of a clutch to be fertilized by one cock, the rest by another. (See Fabricius in Adelmann,^ pp. 191 ss.). Intraovarian fertilization was actually advocated in 1924 by Iwanow (see Hartman*"). The problem of where the spermatozoa remain during this period of great oviducal activity has yet to be solved, but the suggestion of intraovarian fertilization at the present time indicates only ignorance of the histology of the avian ovary and the early history of the bird's egg. (9) The pigeon's egg is over a millimeter greater in diameter after ovulation than the entire follicle was before rupture. This would indicate a rapid resorption of fluid by the yolk at this time and during the condensation of the chalaziferous albumen (Bartelmez,^^ p. 293). There is considerable fluid in the abdominal cavity at the time of ovulation; see §10 of the text. (10) The original in both editions reads: "alba haec vitelli substantia in enixis primum ovis luculentissime observatur." I owe to Dr. Adelmann the interpretation of "primum" indicated in the translation, namely, that Purkinje advises anyone who looks for the latebra for the first time to begin with the readily available laid eggs. As a matter of fact the boundaries of the latebra are clearest in large ovarian oocytes and they rapidly become hazier with incubation. It is pos- sible that Purkinje intended "primum" to signify "newly" laid eggs. (ii) This footnote sounds a bit as if Purkinje had his tongue in his cheek as he replied to a criticism that he had not adequately considered the older literature after the manner of "scholars" who had nothing to write about except what they had read in books. Bellini developed the notion that the primordium of the chick, which he identified with the cicatricula and with the "sacculus amjiii" as well, must be kept away from the air space imtil incubation had begun, otherwise there would be a great accumulation of gas as in the rotten egg! ("Digressio de ovo," "Ovi aere," etc., in his De Motu Cordis,*^ 1693). He be- lieved that the cicatricula passes from the surface into the "viscera" of the yolk on boil- ing as the result of the evaporation of a "liquidum tenuissimum" from the "central cavity." (12) The original reads: "De Baer ... nostram de latebra centrali vitelli ad theoriam de halonibus applicavit." The translation implies that Purkinje was punning and using the original meaning of Oeuipia (a viewing) for his own observations in contrast to the derived meaning, namely, "theory," for von Baer's application. This theory may be found on p. 10 in Vol. I of von Baer's Entwickelungsgeschichte .^ (13) Concentric bands in the yolk were described by William Harvey. Riddle"" in 191 1 was the first to explain the differences betAveen the alternate strata of white and yellow yolk. (14) There is, however, a period in the early growth of the ovarian egg (oocytes from 0.3 to 1.0 mm. in diameter) when the germinal vesicle is not peripheral, although it is always eccentric, so that the polar axis remains obvious (Bartelmez**). (15) This matter was not again carefully studied until 1910 by Maynie R. Curtis" (see translator's note 16). (16) Phillips and Warren"^ have described the process of ovulation in detail. Certain find- ings eliminate oviducal activities as factors in ovulation. Thus it may occur in the absence of the oviduct (Pearl and Curtis),^ or after the petiole has been tied off and the entire follicle has been cut from the ovary and left free in the body cavity (Phillips and Warren^-). This may represent an altered permeability of the vitelline membrane and the taking up of water from the fluid which is abundant in the coelom at this time (see translator's note 9). George W. Bartelmez 85 (17) These "corpuscles" do not appear to be the follicular epithelial cells, which could hardly have been resolved by Purkinje's lenses. In a 9 mm. hen's oocyte on which I have tried to repeat his observations the capillaries were so disposed as to give a punctate appear- ance to the theca interna when viewed horn the inner surface at low magnification. (18) The pendant ovarian follicle is completely enfolded by the rhythmically contracting fimbriae for some time before ovulation as well as during this momentary event itself. (Coste;^* Bartelmez^*). As von Baer says, "so schliirft sich der Eileiter des Huhns die Dotter- kugel ein" (von Baer,^^ vol. II, p. 29). (19) This passage has been translated by A. W. Meyer,^ p. 130. Purkinje wrote: "Ast oh- senationis lacunas opinionum commenla non supplebunl." This aphorism replaces several sentences of the 1825 edition. One of these is worth quoting: "This fluid of the cicalricula of the [laid] egg is by no means to be confused with the vesicle of the ovarian ovulum which I have been describing, although Buffon and Malpighi express themselves in such a way that anyone who confines his attention to words rather than things might easily exclaim [clamitef] about my repeating old stuff [crambem]." Purkinje's crisp and conclusive pres- entation of his discovery was fully accepted by all who understood the matter. (20) A gibe at Carus, Pander, Joerg, and others who still believed that the ovum drops from its ovarian pedicle like an apple from the tree and that the follicular envelopes be- come the shell membranes. It was based on an interpretation of a passage in Aristotle (see Adelmann,^ notes 5-10, pp. 698-700). Meckel von Hemsbaclr''' (1851), von Nathusius''*-^' (1885), and others who saw only what they wished to see, continued this myth. (21) The idea is a development of the Aristotelian tradition which homologized the chalaza with the mammalian umbilical cord. It was developed by Bellini*" and Maitre-Jan^ in 1722 and was revived by Leveille. (22) This is the usual condition in pigeons' eggs, especially at the last formed chalaza (that of the blunt end) which when present at all is always the smaller. Harvey,^" following Fabricius, says that the chalaza at the blunt end is the larger, but Bellini correctly describes the relations in his "Digressio de ovo."^^ (23) This was forgotten and much printer's ink was wasted (see Banclmez," p. 331). 86 Translation from Purkinje REFERENCES 1. Heidenhain, R.: Article on Purkinje, in Allgemeine Deutsche Biographie, XXVI (Leip- zig: 1888), 717. 2. Robinson, V.: "Johannes Evangelista Purkinje (1787-1869)" Sci. Month. 29:216, 1929. 3. Hykes, O. V., and Studnicka, F. K.: "Jan Evangelista Purkyn^" Osiris 2:463, 1936. 4. Eiselt, T.: "Purkyne's Arbeiten" Prager Vrtljrschr. i. prakt. Heilk. Beilage 63:1. 1859. 5. Thomsen, E.: "Cber Johannes Evangelista Purkinje iind seine Werke . . ." Skandinav. Arch. f. Physiol. 37:1, 1919- 6. Studnicka, F. K.: "Joh. Ev. Purkinjes histologische Arbeiten" Anat. Anzgr. 82:41, 1936. 7. Purkinje, J. E.: Symbolae ad Ovi Avium Historiam ante Incubationem (Breslau: 1825). The title page of this work reads as follows: JOAN. FRIED. BLUMENBACHIO EQ. GUELPH. VIRO DE OMNI SCIENTIA NATURALI UNI OMNIUM MAXIME MERITO UNIVERSITATIS GEORGIAE AUGUSTAE DECORI EXIMIO DIE XIX. SEPTLMBRI MDCCCXXV SUMMORUM IN MEDICINA HONORUM SEMISAECULARIA FAUSTIS OMINIBUS CELEBRANTI GRATULATUR ORDO MEDICORUM \'RATISLAVIENSIUM INTERPRETE JOANNE EV. PURKINJE, P. P.O. SUBJECTAE SUNT SYMBOLAE AD OVI AVIUM HISTORIAM ANTE INCUBATIONEM, CUM DUOBUS LITHOGRAPHIS. VRATISLAVIAE, TYPIS UNIVERSITATIS. 8. Purkinje, J. E.: Opera Omnia (Prague: 1918). (Only the 1830 ed. of the Symbolae is in- cluded in these collected papers: p. 195-218.) 9. Cole, F. J.: "Bibliographical reflections of a biologist" Proc. Oxford Bibliog. Soc. 5:169, 1938- 10. Dutrochet, H.: "Histoire de I'oeuf des oiseaux avant la ponte" Jl. de physiq. de chimie etc 88:170, 1819. ii.Coste, J. J. M. C. v.: "Memoiie sur la d^veloppement des mammiferes" Insiitut 1:202, 1833. 12. Wagner, R.: Article on "Ei," in Allgemeine Encyclopadie der Wissenschaften uiid Kiinste, ed. by I. S. Ersch and J. G. Gruber, XXII (Leipzig: 1839), 1. I George W. Bartelmez 8*7 13. Hemvig, O.: "Einleitung iind allgemeinc Liitciaturiibersichl," in Haudhuch dcr Ver- gleichenden und Experimentellcn Entiuickelnngslehrc dcr IVirbellhiere, cd. Ijy O. Hertwig, I. Bd., I. TL, I. Hlft. (Jena: 1906), 1-8-,. 14. Bartelmez, G. W.: "The relation of the embryo to the principal axis of symmetry in the bird's egg" Biol. Bidl. 35:319, 1918. 15. von Baer, K. E.: Ueber Entiuichclungsgeschichte der Thiere—Beobachlung utnl Reflexion, I. TL (Konigsberg: 1828); II. TL, il. Hlft.; (Konigsberg: 1837); ^T. TL, Schlusshft., cd! by L. Stieda (Konigsberg: 1888). 16. Pearl, R., and Surface, F. M.: "Resection and cnd-to-cnd anastomosis of the oviduct in the hen, without loss of function" Amer. Jl. Physiol. 22:357, 1908. 17. Conrad, R. M., and Scott, H. M.: "The formation of the egg in the domestic foul" Physiol. Revs. 18:481, 1938. 18. Cole, F. J.: Early Theories of Sexual Generation (Oxford: 1930). 19. Corner, G. \\'.: "The discovery of the mammalium ovum" Quart. Phi Beta Pi Med. Fratern. 27:76, 1930; also in Lectures on the History of Medicine; a series . . . at the Mayo Foundation . . . 1926-19^2 (Philadelphia and London: ic 19331). 401. 20. Sarton, G.: "The discovery of the mammalian egg and the foundation of modern cm- bryology" Isis 16:315, 1931. 21. Meyer, A. W.: "Essays on the history of embryology" California & West. Med. 35:447, 1931; 36:40, 105, 176, 241, 341, 394, 1932; 37:41, III, 184, 243, 1932. 22. Meyer, A. W.: "Mr. John Hunter on generation" California & West. Med. 43:145, 222, 283, 358, 1935. 23. Meyer, A. 'W'.: An Analysis of the De Generatione Animalium of William Harvey (Stan- ford University, Calif.; London: 1936). 24. Adelmann, H. B.: "The "De ovorum gallinaceorum generationis primo exordio progres- suque, et {^ulli gallinacei creationis ordine' of Volcher Goiter" Ann. Med. Histor. (n.s.) 5:327, 444, 1933. 25. Adelmann, H. B.: The Embryological Treatises of Hieronynius Fabricius of Aquapeti- dente (Ithaca, X. Y.: 1942). 26. Needham, J.: A History of Embryology (Cambridge, Engl.: 1934). 27. Meyer, A. ^V.: Tlie Rise of Embryology (Stanford University, Calif.; London: 1939). 28. von Baer, C. E.: De Ovi Mammalium et Hominis Genesi Epistolam ad Academiam Im- perialem Scientiarum Petropolitanem (Leipzig: 1827). (For "Commentar" on this see Heusinger's Ztschr. f. d. organ. Physik 2: 124, 1828.) 29. Purkinje, J. E.: Article on "Ei," in Eyicycloplidisches Worterbuch der Medicinisclicn Wissenschaften, ed. by C. F. v. Grafe, C. W. Hufeland, H. F. Link, K. A. Rudolphi, and E. v. Siebold, X:io7 (Berlin: 1834). 30. Bernhardt, .: Symbolae ad Ovi Mammalium Historiam ante Praegnationem (Thesis; Breslau: 1834). 31. Coste, J. J. M. C. v.: Recherches sur la Generation des Mammifcres (Paris: 1834). 32. Van der Stricht, O.: "La structure de loeuf de chienne et la genese du corps jaune" Cpts. rd. Assn. d. anatsts. 10:1, 1908. 33. Van der Stricht, O.: "Etude comparce des ovules des mammifcres aux differentes pcriodcs de I'ovogencse d'apres les travaux du Laboratoire d'Histologie et d'Embryologie de rUniversite de Gand" Archs. de biol. 33:229, 1923. 34. Evans, H. M., and Cole, H. H.: "An introduction to the study of the oestrous cycle in the dog" Mems. Univ. California 9:65, 1931. 35. Valentin, G. G.: HandbucJi der Entivickelungsgeschichte des Menschen (Berlin: 1835). 36. Poll, G. S.: Testacea utriusque Siciliae eorumque Historia et Anatome (Parma: 1791-95). 37. Geoffroy Saint-Hilaire, E.: "Organcs sexuels de la Poule: 1" memoire. Formation et rapports des deux oviductes" Mems. Mus. dhistoir. natur. 10:57, '^-'3. 88 Translation from Purkinje 38. Home, E.: "Observations on the changes the egg undergoes during incubation in the common fowl" Philos. Trns. Roy. Soc. (London) 112:339, 1822. 39. Prevost, J. L., and Dumas, J. B. A.: "Memoire sur le developpement du Poulet dans I'oeuf" Ann. d. scis. natur. 12:415, 1827. 40. Dumas, J. B. A.: Article on "Generation," in Dictionnaire Classique d'Histoire Naturelle, ed. by Baron Bory de Saint-Vincent, VII (Paris: 1825), 194. 41. Prevost, J. L.: "De la generation chez la Moule de peintres (Unio pictorum)" Ann. d. scis. natur. 7:447, 1826. 42. Bonnet, C: Considerations sur les Corps Organises (A Neuchatel: 1779; "Oeuvres," III). 43. Hewson, W.: Experimental Enquiries, III (London: 1777). 44. Miiller, J.: Handbuch der Physiologie, I (Coblenz: 1834), 96. 45. Henle, }.: AUgemeine Anatomie (Leipzig: 1841; "Samuel Thomas von Sommerring vom Baue des Menschlichen Korpers," neue umgearb. u. vervollst. Orig.-Ausg., IV). 46. de Vries, H.: "Sur la permeabilite du protoplasma des betteraves rouges" Archs. neerland. d. scis. exact, et natur. 6:117, 1871. 47. Hartman, C. G.: "Ovulation, fertilization and the transport and viability of eggs and spermatozoa," in Sex and Internal Secretions, ed. by E. Allen (2d ed.; Baltimore: 1939). 630. 48. Bartelmez, G. W.: "The bilaterality of the pigeon's egg" Jl. Morphol. 23:269, 1912. 49. Bellini, L.: See translator's note 11. 50. Riddle, O.: "On the formation, significance and chemistry of the white and yellow yolk of ova" Jl. Morphol. 22:455, 1911. 51. Curtis, M. R.: "The ligaments of the oviduct of the domestic fowl" Maine Agricult. Exper. Statu. Bull. no. 176, 1910. 52. Phillips, R. E., and Warren, D. C.: "Observations concerning the mechanism of ovulation in the fowl" Jl. Exper. Zool. 76:1 17, 1937. 53. Pearl, R., and Curtis, M. R.: ". . . On some physiological effects of ligation, section, or removal of the oviduct" Jl. Exper. Zool. 17:395, 1914. 54. Coste, J. J. M. C. v.: Histoire Generate et Particuliere du Developpement des Corps Organises (Paris: 1847). 55. Meckel von Hemsbach, H.: "Die Bildung der fiir partielle Furchung bestimmten Eier der Vogel, im Vergleich mit dem Graffschen Follikel und der Dedicua des Menschen" Ztschr. f. wissnschft. Zool. 3:420, 1851. 56. v. Nathusius, W.: "Besteht eine ausnahmslose Regel iiber die Lage der Pole des Vogel- eies im Uterus im Verhaltnis zur Cloakenmiindung?" Zool. Anzgr. 8:415, 1885. 57. v. Nathusius, W.: "Uber die Lage des Vogeleies im Uterus" Zool. Anzgr. 8:713, 1885. 58. Maitre-Jan, A.: Observatioyis sur la Formation du Poulet (Paris: 1722). 59. Harvey, W.: The Works, tr. by R. Willis (London: 1847), 213. DESCRIPTION OF PLATES PLATE 1 Fig. 1. A small piece of a vitelline membrane with adherent cicatricula in the center of which the clear porus is to be seen. Fig. 2. The same magnified by a lens; the vesicle in the poms is obvious. Fig. 3. The cumulus of the cicatricula viewed from the side toward the interior of the yolk; on its summit is a minute crater, the inner opening of the porus (interna port apostomosis). Fig. 4. The same cumulus of the cicatricula seen from the other side. Fig. 5. The cumulus spread out in the bottom of a vessel after the water has been removed, so that the porus is expanded and shows the vesicle at its center. Fig. 6. The vesicle brought more clearly to view after half of the substance of the colliculus had been removed. Fig. 7. The vesicle after the cicatricula was turned over. Profile view. Fig. 8. The vesicle attached to a portion of vitelline membrane. Fig. 8>1<. A part of vitelline membrane with traces of the zonule and of the torn vesicle. Fig. 9. Vesicle and colliculus cut through the middle. Fig. 10. Magnified cicatricula of a laid egg with the double circle of the blasto- derm and the whitish nucleus in the center. Fig. 11. The same cicatricula with the internal circle of the blastoderm re- moved, so that the fluid cavity with the uncovered nucleus and the farinaceous granules may be seen. Fig. 12. Immature ovarian ovule in which a trace of a cicatricula is to be discerned. Fig. 13. Same cicatricula removed from the ovule hanging to a bit of vitelline membrane. Fig. 14. An even smaller ovarian ovule with a suggestion of the vesicle. Fig. 15. The same containing the vesicle cut open and magnified. Fig. 16. The yolk divided by a perpendicular cut beginning at the cicatricula. In the center the latebra containing a white fluid; from it a canal is continued to the cicatricula. Fig. 17. Yolk cut horizontally; the varicolored strata of yolk material are visible. Fig. 18. Portion of yolk with canal to cicatricula. Fig. 21. A yolk with first stratum of albumen from which the chalaziferous membrane arises; attached to it are the first beginnings of the chalazae which are found above and below it among the folds of the oviduct. Fig. 22. Shows the lamellar structure of the albumen surrounding the yolk. Fig. 22*. A part of the chalaza cut off, within it a canal is seen produced by the spirals of the internal strand of the chalaza. Fig. 23. Part of a vitelline membrane with attached chalaziferous membrane continued into the chalaza [see plate 2]. Fig. 24. Yolk with its chalazae and the whitish girdle of Vicq d'Azyr. 1:90] Plate Isilh: bci I. D. Ciiison in Breslaii. C90] n PLATE 2 Fig. 19. Shows the oviduct of a hen with infundibulum, uterus, and part of vagina. a. Uterus great with egg. b. Vagina, c. Muscular fibers which spread out to the vagina, to the sides of uterus, and to the inferior mesometrium. ddd. Inferior mesometrium; muscular fibers are to be seen in this and vessels running parallel with oviduct. e. Fimbriated edge of infundibulum. /. Mouth of infundibulum. g. Ligament of infundibulum, attached to root of penultimate rib on left side. h. Cut-off part of air sac attached to ligament. Fig. 195^. Posterior part of oviduct where it passes into isthmus with the at- tached mesometrium, composed of interwoven muscle fibers, nerves, and vessels. Fig. 20. Fimbriated part of infundibulum composed of inter^voven muscular network (magnified). Fig. 25. Part of shell membrane whose margin shows the "woolly" structure. Figs. 26-27. ^''^s °f shell with underlying membrane while it is being formed by crystallization. Fig. 28. Elucidating the form of the oviduct. i [192] /.■///), bei /. U. Giiiion in lircflaii. Plaie 2 THE LOCALIZATION OF LIPIDS IN CYTOPLASM By R. R. BENSLEY X From the DEPARTMENT OF ANATOMY UNIVERSITY OF CHICAGO, CHICAGO, ILLINOIS THE LOCALIZATION OF LIPIDS IN CYTOPLASM IT HAS BEEN obvious for somc time that the usual histological methods for the detection of intracellular fat are as a rule successful only in displaying the fats present in the form of discrete globules and leave a large portion of the lipid content undetected. It is even true that cellular organs which display no visible fat may yield a considerable quantity to chemical methods of analysis. This fact has been recently emphasized by the work of Grafllin,^ and of Marble, Grafflin, and Smith/ who made coordinated histological and chemical investigations on the fat content of the guinea-pig liver. The animals used by these observers were kept in a well-fed condition by having food always available, and showed a substantial content of lipids (1.94-3.16 per cent of wet weight equivalent to 6.5-10.5 per cent dry weight), yet the microscopically demonstrable fat was small in amount and irregular in distribution. The majority of hepatic cells were devoid of fat. Pathologists will recall that Rosenfeld* showed long ago that the fat con- tent of kidneys containing little or no visible fat yielded on chemical analysis as much as, or even more than, those in an advanced state of fatty change. Notwithstanding these facts and the warning of Leathes and Raper* that a clear solution might yet contain a considerable amount of fat, histologists have continued to accord to histological methods an undeserved measure of confidence for the detection and localization of lipids. Among the latter the phospholipids have particularly captured the interest of investigators and, one after the other, the visible components of protoplasm, such as mito- chondria, Golgi apparatus, secretion granules, etc., have been identified as phospholipid in nature. Experimental cytologists, on the other hand, have paid little attention to these pseudomicrochemical studies and have in general assumed that the phospholipids were either dispersed in the cytoplasm or disposed on cellular or intracellular surfaces. Since the publication of the Overton theory of the lipoid nature of the semipermeable membranes this point of view has been increasingly popular. The studies of Gorter and Grendel,^ who determined that the amount of lipid extractible from erythrocytes was just sufficient to form a bimolecular layer of lecithin on the cell surface; the observations of Mudd and Mudd' on the ready entrance of erythrocytes into an oil layer with which they were brought into contact; and those of Chambers' on the pene- tration of oil droplets into marine eggs denuded of their ordinary coverings and on the repair of broken cell surfaces— all support the idea that the surface of the cell either consists of or contains lipids. The discovery by Bensley and Hoerr^ that mitochondria of the hepatic cell could be separated from other cell components for analysis by diflcrcntial [97 ] gS Cytoplasmic Localization o£ Lipids centrifugation of suspensions in saline solutions opened up a new line of approach to the problem of the composition of intracellular structures, and in particular the distribution of lipids in cytoplasm. The possibilities of this method were further extended when Claude" announced the successful re- covery, by high speed centrifugation, of a submicroscopic particle from filtered saline extracts of chick embryo. This fundamental discovery, that protoplasm contained particles of complex composition of a size too small to be seen under the microscope except by the use of the dark field, not only clears the way for the investigation of the chemical nature of these constituents of protoplasm but also reveals the nature of the pitfalls which await the bio- chemist in his attempts to separate the constituent proteins of cells. These particles, both microscopic and submicroscopic, have considerable stability within certain pH ranges and electrolyte content, and the answer to the ques- tion as to whether the extract contains only the mobile protein dissolved in the cell water or, in addition, in a state of suspension, the special proteins of submicroscopic particles, mitochondria, nuclei, and other structural ele- ments, can be achieved only by meticulous study at each stage of differential centrifugation with microscopic and chemical examination of each precipitate obtained. The extracts of liver examined by Plosz^° and Haliburton" un- doubtedly contained both mitochondria and submicroscopic particulates. The extracts of the liver made by Luck" using 5 per cent saline at a pH of 5.0 were probably free from particulates but contained some of the structural proteins and nucleoproteins which are insoluble in 0.85 NaCl at pH 6.8, but are soluble in higher concentrations of salt even at relatively low pH. It is unfortunate that Claude, misled by an incorrect estimate in the litera- ture of the size of mitochondria should have identified the submicroscopic particles with mitochondria. Mitochondria maintain a certain uniformity of size and type for a given species and physiological state and do not grade in size down to invisibility. On the other hand, the possibility of fragmentation of mitochondria into submicroscopic elements in saline solution, foreign in composition to the cell water, should not be overlooked. These relations will be fully discussed in papers now in preparation dealing with the composition of the particulates. For the purposes of this article it is sufficient to mention that both mitochondria and the submicroscopic particles have a higher per- centage of fat than that found in the intact cell. The writer's report on the fat content of mitochondria (Bensley") must now be revised, since the dis- covery of the submicroscopic particles and of the particulate nature of intra- cellular glycogen (Lazarow") makes it certain that the original preparations were contaminated to a varying degree with the other two products. Further- more, the phospholipid determinations were far too low, because of the use of the unreliable acetone precipitation method. Mitochondria, separated at centrifuge speeds too low to permit the pre- cipitation of either glycogen or submicroscopics, have a narrower range of variability in fat content than previously reported. The total fat amounts to R. R. Bensley qq 32-38 per cent of the dried product, with a great many preparations yielding about 34 per cent. The phospholipid content of this fat, computed as lecithin from phosphorus determinations, is from 45-58 per cent by weight of the whole fat, and corresponds well with the ratio lecithin x 100 ^^^ ^^^ whole liver. f^t The submicroscopic particles other than glycogen have a higher fat content, ranging from 42-51 per cent of dry weight with about the same or slightly higher percentage of phospholipid. Both particles also contain sterols, and, as pointed out by Claude, the extracted lecithin dissolved in glacial acetic acid gives a positive reaction with Schiff's reagent for aldehydes, indicating a con- tent of acetal-phosphatide. The fat content of the liver in different nutritional states varies within wide limits, but rarely rises above 20 per cent of the dry weight. Mitochondria and submicroscopic particulates therefore contain a much higher percentage of their dry weight in the form of fat than the whole liver. This indicates that some other portion of the cell contains less. It may be recalled at this point that in a series of splendid and significant researches Bungenberg de Jong'^ has explored the conditions of precipitation in lecithin emulsions and in mixtures with known content of lecithin, protein, fat, fatty acids, and various electrolytes. The precipitations observed under these conditions he termed "coacervates." These observations and experi- ments have received little consideration from cytologists, who did not per- ceive that they had a bearing on the structure of protoplasm. Others brushed them aside with the statement that it was just a new name for flocculation, without perceiving that for the first time a quantitative study of flocculation had been made which gave us an insight into the complex composition of particles visible in cells or separable from them. Bungenberg de Jong's speculations are fully confirmed by the discovery of the true composition of mitochondria and of the cytoplasmic submicroscopic particulates of Claude. The experiments of Bungenberg de Jong definitely raise the question as to whether lecithin and fats can exist as molecularly dispersed substances in cytoplasm, a question already answered tentatively in the negative for lecithin by McLean and McLean^" and by Bloor." This problem can now be solved by the use of the methods of investigation introduced by Bensley and Hoerr and extended by Claude. In some types of cell, as for example the liver, the various particulates can be removed seriatijn, examined chemically, and the residual suspension in turn explored for its contents of proteins and lipids. In a long series of investigations directed principally to the problem of the composition of the particulates in the liver cell of the guinea pig my as- sistant Dr. Lazarow and I have been impressed with the fact that, as the particulates are removed by mechanical means, both the protein content and the lipid content of the supernatant solution are reduced, but that the lipid diminishes at a more rapid rate than the protein. A series of experiments loo Cytoplasmic Localization o£ Lipids were therefore undertaken to determine what proportion of the total pro- teins and Hpids respectively were tied up in those little packets which we call the submicroscopic particulates, in the mitochondria, and in the struc- tural proteins of the cell, and also what portion might be considered to be free in the cell water. At this point I wish to make it clear that we have no assurance at present that the submicroscopic particulates constitute a group uniform in composi- tion. Until all the arts of the biochemist are exhausted in the effort to sepa- rate them into uniform categories we shall suspect that many different kinds of particles may be included in this group. The cell emulsions for this study were prepared by the methods previously described. The animals were stunned, quickly bled, and perfused with 0.85 per cent solution of sodium chloride buffered to pH 6.8 with phosphate buffer and cooled to 2° C. Then the livers are removed and homogenized with three times their weight of the same saline, by forcing through gauze and then through bolting silk. The resulting emulsion was centrifuged at 1,900 r.p.m. for 15 minutes to remove intact cells and cell debris. Most of the free fat rises to the surface of the centrifuge tube and may be skimmed off, but since the fat globtiles in the suspension become surrounded with a membrane of surface- denatured protein some may be carried down with the mitochondria and cell debris. For this reason livers that contained few visible fat globules were selected for these studies. The supernatant from the first centrifugation con- tains mitochondria, submicroscopic particulates, and various substances in solution. A second centrifugation for 15 minutes at the same speed removed most but not all the mitochondria. The supernatant was next submitted to a 15-minute centrifugation at 6,000 r.p.m. A new copious precipitate appeared, which is a mixture of mitochondria and submicroscopic particulates. A final centrifugation of the supernatant at 12,000 r.p.m. yielded a clear supernatant and a transparent cherry-red pellet (see also Claude^^) at the bottom of the tube. If the liver contains particulate-glycogen, there will be below the red pellet a transparent colorless layer of glycogen. Three and one- half hours at 12,000 r.p.m. usually suffices to clarify the solution completely, although one must keep in mind that there may be particulates of even smaller size in suspension which could be sedimented by higher rotational speeds. However, at the speed used we are well below the gravitational field at which Wyckoff and his associates recovered their macromolecular substances. Equivalent volumes of the original suspension, before and after centrifuga- tion at the speeds indicated, were analyzed for total protein, total fat, and lecithin content as follows". The suspension was precipitated by the addition of about three volumes of absolute alcohol. The precipitate, recovered by centrifugation, was trans- ferred to a weighed pyrex extraction thimble and extracted for 48 hours with boiling alcohol, then an additional 24 hours with ether. The extracts were added to the original supernatant from alcohol precipitation, and the whole R. R. Bensley lOl evaporated to dryness on a water bath under a continuous stream of carbon dioxide. This extract contained, in addition to the lipids, neutral salts and organic extractives. The residue was extracted with dry chloroform, filtered into a small weighed flask, and evaporated to dryness in a stream of carbon dioxide. The flask and contents, dried to constant weight in vacuo, gave the total lipids. The fats were then dissolved in alcohol, the solution transferred to a volumetric flask and diluted to volume, after which an aliquot was taken for phosphorus determination, from the results of which the total phospho- lipid content, estimated as lecithin, was computed. TABLE 1 Volume of emulsion used, cc. Protein, mg. Fat, mg. Lecithin, mg. Experiment i Whole emulsion Cells removed Particulates removed Experiment 2 Whole emulsion Cells removed Particulates removed Experiment j Whole emulsion Cells removed Particulates removed 25 25 25 25 25 25 50 50 50 577-1 377-8 252.0 386.4 273-4 225-9 717. I 507.0 210.7 141. 2 81.4 10.5 90.5 58.5 8-7 1 1 8 . 2 56.7 1 1. 5 98.25 57-25 0.66 55-40 35-12 3-53 68.00 29-57 1.85 The phosphorus determinations were made by a modified Fiske and Subar- row method, for the details of which I am indebted to Dr. Grafflin of the Harvard Medical School. Table 1 shows the results of these analyses in three separate experiments. The striking feature of these analyses is the almost complete disappearance of fats and phospholipids from these emulsions on long-continued centrifuga- tion, indicating that these substances are associated in the intact cell with stable structural elements, and are present only in traces in the intcrgranular medium. Since the liver cell is a somewhat specialized form of the cell, a similar ex- periment was carried out with the testis of the guinea pig. Four adult guinea pig testes were freed from their capsules and homogenized, as described above, with buffered saline. The emulsion was transferred to a 50 cc. vohmietric flask, diluted to the mark, and thoroughly mixed. Twenty-five cc. were taken for analyses for total fats and phospholipids and the remaining 25 cc. centrifuged for 3I/2 hoias at 12,000 r.p.m. The results are expressed in table 2. In this experiment the loss in phospholipid content as a result of high speed 102 Cytoplasmic Localization o£ Lipids centrifugation is directly comparable to the results obtained in the liver. The percentage loss of other fats is less than in the case of the liver and the ratio of residual fat to residual protein is higher. The difference in the completeness of the removal in the different experi- ments is probably due in part to small globules of fat surrounded by protein haptene membranes failing to precipitate or to rise in the process of centrifuga- tion, and to carry-over of part of the precipitate in pouring off the supernatant because of loose packing in the centrifuge tube. In spite of the completeness of removal of dispersed fats and phospholipids in these experiments, I do not wish to convey the impression that all these substances are contained in mitochondria and submicroscopic particulates. TABLE 2 Quantity Protein, mg. Total fats, mg. Lecithin, mg. Whole emulsion 189.6 78.1 79.8 15-7 35-3 0. 26 Cells and particulates removed The neutral fats, sterols, and sterol esters occur also as discrete globules which, however, contain no phospholipid. The structural proteins of the cell remain- ing undissolved after successive extraction of the cell with normal saline, 10 per cent saline, and 0.005N ammonia in distilled water still contain ap- proximately 25 per cent of their dry weight in the form of lipids. The active participation of lipids must be suspected in the formation of intracellular droplets of a transient sort such as secretion granules, in particu- lar those which brown with osmic acid, such as the zymogen granules of the stomach and pancreas, and artificial droplets such as the neutral red granules of Parat. Nencki and Sieber^" found 10 per cent of lecithin in the crude pepsin obtained by dialysis against distilled water from clear gastric juice obtained by sham feeding from a gastric fistula. Summary It has been found that nearly all of the invisible lipids of the cytoplasm of the hepatic cell of the guinea pig are contained in the particulate components of the protoplasm and in other structural elements, and that the interparticu- late cell water, while containing much protein, contains a negligible quantity of lipids. REFERENCES 1. Grafflin, A. L.: Anat. Recrd. 77:473, 1940. 2. Marble, H.; Grafflin, A. L., and Smith, R. M.: Jl. Biol. Chem. 134:253, 1940. 3. Rosenfeld, G.: Ergbne. d. Physiol., I. Abtl. (Biochem.) 2:50, 1903. 4. Leathes, J. B., and Raper, H. S.: The Fats (2d ed.; London: 1925). 5. Gorter, E., and Grendel, F.: Jl. ExjDcr. Med. 41:439, 1925. R. R. Bensley 103 6. Mudd, S., and Mudd, E. B. H.: Jl. Expcr. Med. 43: 127, 1926. 7. Chambers, R.: Amer. Naturlst. 72:141, 1938. 8. Bensley, R. R., and Hoerr, N. L.: Anat. Recid. 60:449, 1934. 9. Claude, A.: Science (n.s.) 91:77, 1940. 10. Plosz, P.: Pfiiigers Arch. f. d. gsmt. Physiol. 7:371, 1873. 11. Haliburton, W. D.: Jl. Physiol. 13:806, 1892. 12. Luck, J. M.: Jl. Biol. Chem. 1 15:491, 1936. 13. Bensley, R. R.: Anat. Recrd. 69:341, 1937. 14. Lazarow, A.: Science (n.s.) 95:49, 1942. 15. Bungenberg de Jong, H. G.: Protoplasma 15:110, 1935. 16. MacLean, H., and MacLean, I. S.: Lecitliin and Allied Substances (2d ed.; London: 1927). 17. Bloor, W. R.: in A Textbook of Biochemistry, ed. by B. Harrow and C. P. Sherman (Phila- delphia and London: 1935), 584. 18. Claude, A.: Cold Spring Harbor Sympsa. Quantit. Biol. 9:263, 1941. 19. Nencki, M., and Sieber, N.: Gaz. lekar. 21:422, 455, 482, 500, 1901; Hoppe-Seyler"s Ztschr. f. physiol. Chem. 32:291, 1901. THE SOURCE OF EQUINE GONADOTROPHIN By H. H. COLE AND HAROLD GOSS From the DIVISION OF ANIMAL HUSBANDRY UNIVERSITY OF CALIFORNIA, DAVIS, CALIFORNIA I i THE SOURCE OF EQUINE GONADOTROPHIN THE BLOOD and tissues of a mare at the 70th day of pregnancy contain 1 to 4 million I.U. of gonadotrophin. This paper relates to the source of this hormone. Catchpole and Lyons^ considered this question and concluded that the chorion produced the hormone on the basis of their findings: 1. The mare pituitary gives very little histological or physiological evidence of increased secretory activity at the time when the hormone level is increasing rapidly. 2. Large amounts of hormone are stored in the endometrium. To accept the view of hypophysial origin one must assume that the hormone is selectively accumulated in the fertile horn of the uterus. 3. One case of early pregnancy was found in which hormone was present in the chorion but not in the blood or endometrium. This, they felt, gave evi- dence that hormone was first formed in the chorion from which it later passed to the endometrium and maternal blood. 4. The endometrium in contact with the chorion is much richer in gonado- trophin than is the endometrium of the infertile horn, though marked proges- tational proliferation occurs in both. In 1937 we had occasion to test the potency of the endometrium of several mares and were struck by the cup-shaped structures distributed over that part of the endometrium in contact with the chorion. Testing of these cups re- vealed an extremely high potency as compared to the intervening endo- metrium. As we did not take samples of these organs for histological study or attempt to make quantitative studies on the amount of hormone in vari- ous tissues, publication was withheld until more complete data were available. In 1941 we had an opportunity to obtain material from four grade Welsh pony mares at known stages of pregnancy, and this material forms the basis of the present report. The high concentration of gonadotrophin in these cups led us to suspect that they might be the source of the hormone. The present studies were initiated to determine the validity of this view. We shall first describe these structures briefly and then present the data comparing the potency of these cups with other tissues of the mare. Description of the Endometrial Cups Most of the early workers studying the changes in the endometrium of the pregnant mare failed to observe these structures because of lack of material during the stages in which they are present. Schauder^ has described them, however, in considerable detail and we shall only attempt to add to his de- scription. He expressed the view that the hippomanes of the horse are derived from these cups— an opinion which we can confirm, in part at least, for in mare 4, sacrificed at the 105th day of pregnancy, the cups appear as pendulous structures from the allanto-chorion extending into the allantoic cavity, 1:107] 1 o8 Source o£ Equine Gonadotrophin Three mares were sacrificed between the 6oth and '/oth days of pregnancy. In all instances the cups appeared as elevated oval structures varying in maximum diameter from a few millimeters to 3 cm. They are distributed, sometimes in a circular arrangement, over only a portion of the endometrium in apposition with the chorion. From a cursory examination one might con- clude that they were cotyledons provided as a special means of attachment be- tween the maternal and fetal membranes. Actually there is no attachment of these membranes at this point; rather, the membranes are separated, due to the fact that the cups are filled with a yellowish-brown, waxy, or glue-like material. In mare 4, killed at the 105th day, this material was so thick that it could be peeled out with its almond shape intact. In the other cases, however, it adhered like thick glue to the cup or to the chorion. In our opinion, adher- ence of this material to the chorion explains the one case (K5) cited by Catch- pole and Lyons^ in which the chorion gave a positive response while the endometrium was still negative. We find the potency of the endometrial-cup secretion to be 10 to 30 times greater than that of the cups. Schauder has well illustrated the gross appearance of these cups. Histologically the cups have a unique appearance in contrast to the remain- ing endometrium, as also does the chorion overlying these cups. As Schauder has pointed out, the chorion at these points is devoid of villi (pi. 1, figs. 1 and 2). Mossman,^ speaking generally, states that chorionic villi are frequently lacking in areas where the chorion does not come in contact with the endo- metrium. No doubt, the separation of maternal and fetal membranes by the endometrial-cup secretion explains the lack of both chorionic and endometrial villi at these points. The epithelial cells of the chorion in the cup region are of the tall columnar type and give evidence of greater secretory activity than does the chorion in other regions. The lumens of the uterine glands are greatly enlarged, having a cross- diameter several times that of the glands in other portions of the endometrium. The lumens are filled with secretion and the pouring out of this material accounts in part for the endometrial-cup secretion. Between the Goth and 70th days the necks of the glands are in a poor state of repair; the basal glands have a much healthier appearance (pi. 1, fig. 2). At the 105th day the uterine glands are the conspicuous feature of the endometrial cups (pi. 1, fig. 1 and pi. 2, fig. 4). The glandular epithelium is extremely low and still the glands appear to be secreting rapidly (pi. 2, fig. 4). The tissue between the uterine glands in the cup area does not present a constant picture. In mare 2, sacrificed on the 63d day of pregnancy, many small cells with scanty cytoplasm are packed between the basal uterine glands (pi. 2, fig. 3). In the upper part of the cup there is evidence of autolysis of both glandular and interglandular tissue (pi. 1, fig. 2). In the other two cases (mares 1 and 3), sacrificed between the 6oth and 70th days, extremely large polyhedral cells are present between the glands, though some of the smaller cells similar to those seen in mare 2 are also present (pi. 2, figs. 5 and 6). As the cup from Cole and Goss loq mare 2, selected for study, is much smaller than those selected from the other mares, it is likely that this case represents an earlier stage. Cups of various sizes are found side by side at this stage of pregnancy. We saved a minimum of mate- rial for histological study because of our interest in hormonal concentration. This development of the interglandular tissue is comparable to the decidual responses of the endometrium during pregnancy in other forms. Plate 1, figure 1, gives evidence, in truth, that the superficial part of the cup is sloughed, as the outline of the cellular debris is still evident. This sloughing begins early in pregnancy, as plate 2, figure 6, shows; the epithelium of the mucosa is already completely lost. In plate 1, figure 2, the uterine epithehum is still intact at the lips of the cup but completely lacking in the central portion. Perhaps one is justified in referring to this as a decidual response, even though the tissue is shed prior to parturition. Plate 1, figure 1, shows the pendulous character which the cups assume at the 105th day of pregnancy. These bodies extending into the allantoic cavity are, no doubt, an early stage of the pendulous hippomanes. Concentration of Gonadotrophin in Endometrial Cups and Other Selected Tissues of the Mare Preparation and Assay of Tissue Extracts. Only a small amount (37 mg.) of endometrial secretion was taken for assay from the endometrial cups of mare i . This was dried in a vacuum desiccator, again weighed, and then taken up in 10 cc. of sterile saline for assay. This secretion was found to contain 29.5 per cent solids. The amount of secretion in this mare was very small. We did not, however, attempt to remove it quantitatively, for we did not know of its im- usual activity prior to this time. The endometrial secretion in mares 2, 3, and 4 was much more abundant. These thick mucus-secretions were carefully scraped from the cups and weighed, then taken up in either 100 or 200 cc. of sterile saline, forming a uniform opalescent solution. A small aliquot was removed from each solution for assay and the remainder frozen at once. The endometrial cups, the endometrium between the cups, and the chorion were weighed and ground with sand under sterile saline. The mixture was centrifuged and the residue reground and extracted as before. This process was repeated three or four times and the extracts combined. The volume of the extract was noted, a sample removed for assay, and the remainder frozen. The pituitary was extracted by repeated giinding with sand and 2 per cent pyridine solution, separating the extract by centrifuge (Fevold^ suggested using 2 per cent pyridine for extraction of pituitary tissue). To the combined extracts were added 6 volumes of 100 per cent acetone. The precipitated material, after standing at 2° C. over night, was removed by centrifuging, and taken up in 20 cc. of sterile saline for assay. The tissues, with the exception of the pituitaries, were assayed by the method of Cole and Erway.^ The pituitary extracts were assayed by the weight method of Cole and Saunders" suggested for mare serum, with the exception that the no Source o£ Equine Gonadotroph in injections were made on three consecutive days with autopsy 1 20 hours after the initial injection. This is an arbitrary procedure,— attempting to make the assays as comparable as possible to those of the other tissues. Inasmuch as the nature of the gonadotrophic response of pituitary tissue differs from blood and embryonic tissues, it is impossible to make a direct comparison. For ex- ample, 64 mg. of fresh pituitary tissue of mare 4 gave an average ovarian weight of 14.5 mg., in a group of four rats and there was no evidence of develop- TABLE 1 Amount of Gonadotrophin (I.U. per gm.) in Fresh Tissues and in Endometrial-Cup Secretion of Pregnant Mares Days pregnant at autopsy Crown- rump length of fetus in cm. I.U. per gm. of fresh tissue Mare No. Endometrial- cup secretion Endometrial cups Endome- trium be- tween cups Chorion Blood serum Pituitary* I 62 6.2 70 , 000 6,750 380 10 200 500 2 63 7-7 144,000 4,100 200 20 120 250 3 67 8.1 314,000 I 2 , 900 750 250 170 250 4 105 19.7 50 , 000 4,500 17 5 70 50 • The assay of pituitaries is not in terms of I.U. but on an arbitrary basis. See text for details. ment either of the ovaries or of the accessory reproductive organs, while 128 mg. gave an average ovarian weight of 38.2 mg. with estrous uteri and vaginal smears in all animals. With mare serum one regularly obtains a clearcut ovarian and accessory organ response with one-sixth of the dose necessary to produce ovaries weighing 38 mg. Gonadotrophin m Various Tissues. Table 1 gives the amount of gonado- trophin per gram of fresh tissue. The endometrial cups contain 50 to 900 times more hormone per gram than the chorion and 17 to 260 times more than the endometrium between the cups. The endometrial-cup secretion contains 10 to 35 times more hormone per gram than do the endometrial cups. The slight activity of the chorion could easily be explained by the adherence of a small amount of this secretion to it. Table 2 gives the total amount of gonadotrophin in various tissues. One endometrial cup with the chorion attached was saved for histological study in each case and this was not taken into account in the computation. Thus, the figures on the amount of hormone in the endometrial cups, endometrial- cup secretion, and chorion are somewhat low— approximately 1-5 per cent of the total tissue was saved for histological study. z D .J < O I— I < U m H 2 I— I b O 0. O a f- o c < z o O o H Z D O S < -9 h O Amount of hormone, l,ooo I.U. >-o >J-i r-5 'J- O d d d d *3 .— *-t Oh O o « c o hH <-o >-H E 2 "> '"' ^ •"■ *-• ■< *" t: «J'5 o O O o o O O o * B o ct r-- r- < JO. cT m" S TJ O I 5^' • o o o o ^ o o o o W o-oSS o o o o o^ o 6 o t: «j"3 d d «j-. d .2 < -§° 'V, o *-> i°.2& o >~D <~o \D ^>D 4 d ci " cs d Tt- t: '"■^ C c ■ a « 2 3 3^ 0" i°g8 ^D o t~o 6 .2 t* r-5 oo oo ^o o d 1— ( t Tt- cl < " ^ JD c c • /^ 5— ="■ g o S o S 1- o Cf < 2 °. eg- O (J -a c (M 1— 1 ^ ^o W " oo d 4 < " " iTD D. c ° S o -t ^ -t 3 r- <^ c^ w^ 4 .2 o < iS- 1" 4-^ n «j" . ^ f^ ,—1 c §•^5 5 E .2 t* ++ (N CO ^ c %■- CI CO r^ w-i \D ^ \D O c 6 2 u u rt ■^ ^ (H r- ■d 3 -Si -a « 'c o al H' -^ qj U J= m £ c« CO 1> > 00 O E 4> 3 O U CO V -n T3 o o E -o CO t_ ol 0 fi *~t n o e R r( *-> o HZ 112 Source o£ Equine Gonadotrophin Discussion With the additional facts which we now have at hand, let us again consider the question of the source of equine gonadotrophin. The pituitary, chorion, and endometrium can be considered to be the most likely sources. Pituitary. If the pituitary secretes the hormone, one must consider the possi- bility that gonadotrophin is selectively taken up by endometrium and more particularly by the endometrial cups, as the hormonal concentration of the cups is much greater than that of the blood. Then one must assume that the cups in turn secrete the hormone into the space between the maternal and fetal tissues in a still more concentrated form. In other words, the cups would be serving as excretory organs. Such a sequence of events seems to us extremely unlikely. If the pituitary were the origin of the hormone then one must as- sume an unusual activity for this gland. For instance, in mare 3 there is a total of 2.4 million I.U. of hormone accounted for, whereas the pituitary con- tains only 400 I.U. (table 2). This amount of hormone has accumulated in the tissues over approximately a 25-day period, as the hormone first appears in the bloom stream at about the 42d day of pregnancy. Thus it would be neces- sary for the pituitary to secrete its total hormonal content 240 times daily to account for the total hormone. It is true, of course, that some complications arise in attempting to compare hormonal concentrations in the pituitary with those in other tissues, as we have previously pointed out. Part of these differences in action may be explained by the finding that the pituitary, unlike mare serum, contains antagonistic substances and further that the pituitary gonadotrophin is excreted in the urine (Evans and co-workers^. The differ- ences in biological action of gonadotrophin obtained from the pituitary and from serum or placental tissue might be used for a further argument against the view of pituitary origin of serum gonadotrophin. Chorion. We have reviewed above the arguments used by Catchpole and Lyons^ for the chorionic source of the hormone. The fact that the chorionic epithelium overlying the cups gives evidence of special secretory activity could be used in further support of this view. A study of hormonal concentra- tions in the tissues, however, does not, in our opinion, support this theory. The endometrial cups contain from 50 to goo times more hormone per gram than the chorion in the 4 mares of the present study (table i) . The endo- metrium in the cup region is separated from the chorion by the mass of edo- metrial-cup secretion (pi. i, fig. 1). If the hormone is produced by the chorion, one can account for the high activity of the endometrial-cup tissue in two ways: gonadotrophin can diffuse slowly into the tissue from the mass of material which we have referred to as endometrial-cup secretion; or it can pass to the fetal blood, thence to the maternal blood at areas where there is close apposi- tion of maternal and fetal membranes, and then be selectively absorbed from the maternal blood by the endometrial cups. The fact that the hormone is a protein and thus diffuses slowly renders the first possibility less tenable. The Cole and Goss 1 1 ^ fact that practically no hormone is present in fetal tissues (Catchpole and Lyons^) makes the alternative supposition untenable. Endometrium. It is our belief that equine gonadotrophin is produced in the endometrium and for the most part by the endometrial cups. Catchpole and Lyons^ were the first to draw attention to the high concentration of hor- mone in the fertile endometrium. Against the view that the endometrium secretes the hormone, however, they cited the fact that the fertile endometrium is high in hormone, whereas the infertile horn is low, even though both horns undergo progestational proliferation. The finding of specialized structures, the endometrial cups, confined solely to the fertile endometrium and very rich in hormone, removes this objection. Another point which they raised against the secretion of the hormone by the endometrium was that in one early pregnancy case gonadotrophin was found in the chorion but not in the endometrium or blood plasma. In this case, they reasoned, the hormone had been produced in the chorion but had not as yet passed in recognizable amounts to the other tissues. As previously pointed out, this case may be ex- plained by assuming that the gluelike endometrial-cup secretion had adhered to the chorion rather than to the endometrium. As this secretion is extremely potent, a few milligrams of it would account for the total activity manifested by the chorion. Even though the assumption of adherence of endometrial-cup secretion to the chorion were false, the finding of hormone in the chorion while it is absent in the endometrium of this mare (K5) is nullified as an argu- ment for chorionic source of the hormone by the fact that in two other cases of early pregnancy which they studied (D31 and D23) relatively large amounts of hormone were present in the endometrium while the chorion contained none. Now that we have considered the objections raised to the secretion of gonadotrophin by the endometrium, what is the positive evidence in favor of this vicAV? 1. The endometrial cups contain more hormone per gram than any other tissue. The relatively minute amount of hormone in the chorion, in fact, almost precludes it as a possible source (tables 1 and 2). 2. The waxy material in the endometrial cups contains more hormone than any tissue of the mare (in mare 3 up to 300 1.U. per milligram of fresh secretion or 600 I.U. per milligram of total solids) and a major part of this material which we have referred to as endometrial-cup secretion certainly is derived from the endometrium. 3. One can account for the coincident high concentration of the hormone in the endometrial-cup secretion and in the blood most easily by assuming an endometrial source. 4. The endometrial cups are specialized structures to which one can at- tribute special secretory activity. 5. Considerable hormone is present in the endometrium between the cups (table 2). This strongly suggests that the hormone is not formed exclusively by the endometrial cups. If one were to assume chorionic production of the 114 Source o£ Equine Gonadotrophin hormone, the endometrium between the cups might be expected to contain more hormone than the endometrium in the cup area because of the close apposition of maternal and fetal tissues and consequently the better oppor- tunity for exchange (pi. i, fig. i). On the contrary, however, the cups contain twenty times or more hormone per gram and the total hormone in the cup is at least three times greater, even though the total weight of the cup tissue is much less than that of the intervening endometrium (tables i and 2). To what cells of the endometrium should one ascribe the secretion of gonadotrophin? Presumably both the glands and the interglandular tissue contribute to the endometrial-cup secretion. Also both tissues differ markedly histologically from the endometrium between the cups. A number of cases sacrificed just prior to and after the appearance of hormone in the blood should provide additional evidence on this point. Summary and Conclusions Quantitative studies on the gonadotrophic concentration in the tissues of four mares, sacrificed between the 62d and 105th days of pregnancy, were made, coincident with histological studies of the endometrium and chorion, in order to elucidate the source of equine gonadotrophin. Distributed over a part of the endometrium in apposition to the chorion, specialized structures are found which we have referred to as endometrial cups. These cups contain 4 to 12 I.U. of gonadotrophin per milligram of fresh tissue and furthermore they elaborate a secretion into the uterine space containing 50 to 314 I.U. per milli- gram of secretion in the fresh state. The histological character of both the glands and the interglandular tissue of these cups differs conspicuously from the endometrium in other areas. We conclude that these endometrial cups are the chief source of equine gonadotrophin. REFERENCES 1. Catchpole, H. R., and Lyons, W. R.: Amer. Jl. Anat. 55:167, 1934. 2. Schauder, W.: Arch. f. Anat. u. Entwcklgsgeschte. 1912:259. 3. Mossman, H. W.: Contribs. Embryol., Carnegie Instn. Washington 26:133, 1937. 4. Fevold, H. L.: Endocrinology 24:435, 1939. 5. Cole, H. H., and Erway, J.: Endocrinology 29:514, 1941. 6. Cole, H. H., and Saunders, F. J.: Endocrinology 19:199, 1935. 7. Evans, H. M.; Simpson, M. E., and Austin, P. R.: Jl. Exper. Med. 58:561, 1933. DESCRIPTION OF PLATES PLATE 1 Fig. 1. Endometrial cups from mare 4 sacrificed on 105111 day of pregnancy. The cups are pendulous at this stage. Note the tremendous enlargement of the lumens of the uterine glands in the cup area. al. c, allantochorion; e.cs., en- dometrial-cup secretion; u.g.c, uterine glands in cup area; u.g., uterine glands outside of cup area, x 7. Fig. 2. Endometrial cup from mare 2 sacrificed on 63d day of pregnancy. The uterine epithelium is entirely absent and there is autolysis of all tissues in center of cup. X 14. \ n-'S] u g. /s ~% ^ .A. ^\ .age lo/.c. V / i \ \ y \ /■ G.C.S. r^^: V '* 2 {^-^ v- £<•.* "<, I • •5a. -^ "M \. -( • ■ < 5M iif^ 53 ,<5 n; ^ ,-f^- -»"''?!'*"^ "-/Tb; ^,. (•^ lP4 » ^"••S**, X 'f*,' '' i"? %.J ft . 'A e' y* '■ » " (J Ms 'M e~^:^::j~^. Plate 2 [119] I 4 ON THE FEMALE TESTES OR OVARIES By REGNER DE GRAAF Chapter XII oiDe Mulieruin Organis Generationi Inservientibus (Leyden: 1672) Translated by GEORGE W. CORNER From the DEPARTMENT OF EMBRYOLOGY CARNEGIE INSTITUTION OF WASHINGTON BALTIMORE, MARYLAND TRANSLATOR'S NOTE I OFFER THIS English version of De Graaf's famous description of the ovaries, for publication in honor of Herbert McLean Evans, in the hope that it will recall to him the happy days when he and I worked side by side in his laboratory, students of the ovary no less enthusiastic than our predecessor De Graaf. Himself, like De Graaf, a distinguished contributor to the physi- ology as well as the anatomy of the reproductive system, he will (I hope) find pleasure in reading this masterpiece of the brilliant young Dutchman. In another place (Corner, 1930) I have discussed the background and the relative importance of the contribution made by De Graaf. It will be suffi- cient to say here that what follows is the first thorough description of the mammalian female gonad and that it proved and established the fact that this organ, like the ovary of birds, is actually an ovary. De Graaf did not, of course, identify the actual ovum; by a very excusable error he assumed that the whole contents of the follicle is the egg. He was, indeed, much surprised when he traced the rabbit's embryos back to the third day after copulation, when they were still blastocysts in the oviduct, to find them so much smaller than the follicle. This discrepancy De Graaf never solved. His name has long been attached to the ovarian follicle, but it might with more justice have been applied to the corpus luteum; for the latter receives in this chapter its first good description, whereas the follicles had been seen previously and the description given of them here is not much more extensive than that of his predecessors. The chapter has, to the best of my knowledge, been translated into modern languages only twice before. The whole book on the female reproductive system is included in the Dutch translation of the complete works of De Graaf, published at Leyden in 1686. Our chapter on the ovaries was trans- lated into English by that remarkable anatomist of Edinburgh, Robert Knox, in 1848. This version, published in an obscure place, is now scarce. The Army Medical Library kindly permitted me to have it photographed. In polishing my own version I have not hesitated to borrow a happy word from Knox now and then, and in return I believe I have corrected one or two errors in his translation. In doubtful cases I have accepted as probably correct the reading of the Dutch translation, made within a few years of the original Latin text. The three plates which accompany the chapter have been photographed from a copy of the first Latin edition in my possession. The portrait of De Graaf has been redrawn by Mr. D. K. Winter from the contemporary engraving which was used as the frontispiece of De Mulierum Orga7iis and of the complete Opera. [1233 REGNER DE GRAAF 1641-1673 CHAPTER XII C 171 r ON THE FEMALE TESTES OR OVARIES THE TESTES of womcn differ much from those of the male as to posi- tion, form, size, substance, integuments, and function, as we are about to describe. Thus, they have not an external position as in men, but are located in the lowest portion of the abdominal cavity about two finger breadths from each side of the fundus of the uterus, to which they are attached by a strong ligament which is called "Vas Deferens" by many anatomists, because they believed that semen was transferred through it from the testes to the uterus; on the other sides they are firmly attached to the peritoneum about the region of the iliac bone by the spermatic vessels, which supply them, and by the membranes with which the spermatic vessels are involved; so that the testes, fixed [^172] on each side, as if suspended, reach about the same level as the fundus of the uterus in the non-pregnant; in the pregnant, however, although they follow the fundus of the uterus to some extent, they do not rise to an equal degree, and thus, the more the fundus of the uterus rises, the farther they are from it, always keeping a lower position. The testicles are not suspended by any cremaster muscle, although some state this opinion, following Soranus. They are located in the interior cavity of the abdomen, in order that they may be nearer the uterus and serve the better and more easily their intended purpose, which will be fully demonstrated below. The testes of women, since they are broad and flattened on their anterior and posterior sides, differ much from those of men, for in their lower part they have a semi-oval bulge, while in the upper part, which the blood vessels enter, they appear more flat than humped, C 173 H so that the testicles when separated from the blood vessels and the ligaments present a somewhat flattened semi-oval form. The surface is more uneven than in males, because on account of the contents it projects unequally here and there, and displays certain small fissures in different places from time to time, due to depression or retraction of its coverings. Moreover, their size varies not a little with age, for in developing girls and [in women] in the flower of their life they weigh almost one and a half drachms, so that they attain a size about half that of the male testis, although in proportion they are wider and more succulent. In the old and decrepit, they are smaller, firmer, and more dried up, and slowly wither more and more, * Numbers in the margin indicate the pages of the original Latin text of 1672. Words in brackets are exphmatory additions by the translator. O25] \d> r^i 126 Translation from De Graaf but never disappear completely; we have observed that the smallest testicles of old women weigh one scruple. In newborn and young [^174] infants they usually weigh from five grains to half a scruple; and therefore are smaller in these than in the very old, although most anatomists say they are larger in infants and gradually diminish with the thymus gland. In exceptional cases, however, the testicles grow to a remarkable size and contain within them such a quantity of fluid that they are dropsical: of which condition Schenck gives many ex- amples in his observations, and Riolan and others, as well. Moreover, the coverings of these testicles differ much from the male, for the latter are enveloped by many tunics, so that although hanging freely, they are protected from all injury; but the former are not altogether deprived of such protection, since they are invested by a tunic peculiar to themselves, called dartos by Galen; which although it is only moderately tough, is not easily removed from the substance C ^75 H of t^^^ testes, for it adheres to them as if continuous with their sub- stance. A membrane arising from the peritoneum covers the upper part of the [male] testicles and the blood vessels supplying them, and the same is generally considered to be true also of the female testicles. Some, however, who neither by boiling [the ovary] nor by any other artifice have been able to distinguish this membrana propria from the peri- toneum, by a different appearance, agree with us that the [female] testes are covered with a single membrane originating from the peri- toneum, and that it seems thicker [than the covering of the male testis] because it is so much more firmly bound and united to the [subjacent] parenchyma that one can scarcely see how it may be sepa- rated from it or divided into many membranes; but since this is not a matter of great importance, let us leave the decision as to the number of these coverings free to everyone. [ 176 ^ When the covering of the testes is removed, their whitish substance is revealed, which in every way differs from the substance of the male testicles, for the former, excluding from consideration certain mem- branes and nutritive vessels, are composed of seminal vessels which if mutually joined to each other, would exceed twenty or even forty ells in length; the testicles of women are not composed of similar vessels and no one, diligent as he may be, can in the least separate them [into vessels]. Their internal substance is chiefly composed of many membranes and fibrils, loosely bound to one another, in the interstices of which are found many bodies which are either normal or abnormal. The normal structures, regularly found in the membranous substance of the testicles just described, are vesicles full of liquor, nerves, and C 177!] nutritive vessels, which [that is, the blood vessels] run to the testes George W. Corner 127 in almost the same way as in males, as we have already described, and course throughout the whole of their substance, and enter the vesicles, within whose tunics many branches end after free division, in just the same way as we have seen happening in the ovaries of fowls com- posed of clustered egg yolks. As to whether the lymphatics found in the testes enter their sub- stance, we have not yet determined with sufficient clearness to ven- ture an assertion, although this seems probable. Those structures, which though normal, are only at certain times found in the testes of women, are globular bodies in the form of con- glomerate giandulae which are composed of many particles, extending from the center to the circumference in straight rows, and are en- veloped by a special membrane. We assert that these globules do not exist at all times in the testicles of females; on the contrary, they are [ 178 ^ only detected in them after coitus, [being] one or more in number, according as the animal brings forth one or more foetuses from that congress. Nor are these always of the same nature in all animals, or in the same kind of animal; for in cows they exhibit a yellow color, in sheep red, in others ashen; because a few days after coitus they are composed of a thinner substance and contain in their interior a limpid liquor enclosed in a membrane, which when ejected with the mem- brane leaves only a small space within the body which gradually disappears, so that in the latter months of gestation they seem to be composed of a solid substance; but when the foetus is delivered these globular bodies again diminish and finally disappear. Finally, the abnormal objects sometimes found in the testes of women are hydatids, calculi, steatomata, and other similar things. [^ 179 ] From what has just been said, everyone will readily gather that it is the vesicles or their contents solely, which the nerves, arteries, veins, integuments, and the other structures normally observed in the testes are designed to serve. These vesicles have been described under various names by Vesalius, Fallopius, Volcher Goiter, Laurentius, A Castro, Riolan, Bartholin, Wharton, Dom. de Marchettis, and others, whose accounts it would be too tedious to repeat here in full; it will not, however, be amiss to quote two of them at the present time, in order that the truth may be confirmed by their words. Fallopius says in his anatomical ob- servations: "I have seen in them indeed certain vesicles, as it were, swollen out with a watery humor, in some yellow, in others trans- parent"; A Castro, also (lib. i, cap. 4: "De Natura Mulierum") , says: "The testes have within them, besides the vessels, certain cavities full [ 180 ] of a thin and watery humor which is like whey or white of egg." Some call these vesicles hydatids, but the celebrated Dr. van Home in his Prodromum preferred to call them ova, a term which since it seems 1 28 Translation from De Graaf to me more convenient than the others, we shall in the future use, and we shall call these vesicles ova as does that distinguished man, on account of the exact similitude which they exhibit to the eggs con- tained in the ovaries of birds; for these, while they are still small con- tain nothing but a thin liquor like albumen. That albumen is actually contained in the ova of women will be beautifully demonstrated if they are boiled, for the liquor contained in the ova of the testicles acquires upon cooking the same color, the same taste and consistence as the albumen contained in the eggs of birds. It is of no importance that the ova of women are not, like those of [^ 181 ] fowls, enveloped in a hard shell, for the latter are incubated outside the body in order to hatch the chickens, but the former remain within the female body during development, and are protected as thoroughly from all external injuries by the uterus as by a shell. But before we proceed farther in their description it must be de- termined whether they are found in animals of all kinds and in what way they differ from hydatids. We may assert confidently that eggs are found in all kinds of ani- mals, since they may be observed not only in birds, in fishes, both oviparous and viviparous, but very clearly also in qtxadrupeds and even in man himself. Since it is known to everyone that eggs are found in birds and fishes, this needs no investigation; but also in rabbits, hares, dogs, swine, sheep, cows, and other animals which we have dissected, those structures similar to vesicles exhibit themselves to the eyes of the dissectors like the germs of eggs in birds. Occurring in the superficial part of the testicles, they push up the common tunic, [^ 182 ] and sometimes shine through it, as if their exit from the testis is im- pending. These ova differ much in animals of various kinds, for we have observed that in rabbits and hares they scarcely exceed the size of rape seeds; in swine and sheep they reach the size of a pea or larger; in cows they often exceed the size of a cherry. It must be noted, however, that in these animals, besides the large ova, lesser ones are found, of which some are so minute that they may scarcely be seen, for age and sexual intercourse cause gieat changes in the ova. In younger animals they are smallest, in older animals they become greater, and after coitus they are changed into the globules formerly described, of which one or more are formed, according to the number of embryos produced by the animal. These ova are so [] 183 ] plentiful that we have sometimes seen twenty or more, filled with very clear liquor, in one testicle. Believing that these conditions are found in all animals dissected as yet by us, we asked the eminent Dr. N. Steno if he would deign to communicate to us what he had observed of the female testes in various other animals, of which w^e had not George W. Corner 1 29 sufficient specimens, or which we had had no opportunity to dissect. He granted our request and generously informed us that in fallow deer, guinea pigs, badgers, red deer, wolves, asses, even in mules, and in other animals he had found ova of diverse sizes. These observations, combined with our own, more than sufficiently confirm the finding of ova in the females of all species. If any one inquires why they are present in the aged and in mules, which are incapable of reproduction, we say only that they are no more serviceable than the uterus, [male] testes and other reproductive organs customarily fovuid in these as [ 184 ] well as in fertile animals; many reasons may be given for their steril- ity, as for instance an improper conformation of their organs, in- sufficiency of the material of the ova for conception, or many other of the possible causes for sterility. Since we have said above that we have sometimes found in the sub- stance of the testicles or in their membranes vesicles of another kind very similar to the ova, it becomes important to cite in this place the chief differences between them and the ova. Vesicles of the other kind, called hydatids, are usually formed w4th a double tunic. The interior layer, although very thin, is by no means difficult to separate from the exterior and the liquid contents is not easily coagulated by boiling. On the contrary, the common coats of the ova are separated from each other with great difficulty and their liquor is coagulated by boiling; hence, whenever we have found in testes which have been boiled some vesicles filled with hardened sub- [] 185 ] stance and others with a liquid humor, we have considered the former ova, the latter hydatids. It must be added that the hydatids now and then are suspended from the membranes of the testicles as if by a peduncle, which as yet we have never found to be the case with true ova. These ova arise and are developed in the testes in exactly the same way as the eggs in the ovaries of birds, inasmuch as the blood flowing to the testes through the nutritive arteries deposits in their membra- nous substance materials suitable for the formation and nourishment of the ova, and the residual himiors are carried back to the heart through the nutritive veins or lymphatic vessels. After the ova acquire their normal size they become invested with numerous tunics or follicles and in these immediately after sexual intercoinse a kind of glandular substance grows up, of which the contents of the globules just described is composed. We shall attempt below to explain the [^ 186 ] purpose for which nature has so arranged [them]. Thus, the general function of the female testicles is to generate the ova, to nourish them, and to bring them to maturity, so that they serve the same purpose in women as the ovaries of birds. Hence, they should rather be called ovaries than testes because they show no 130 Translation from De Graaf similarity, either in form or contents, with the male testes properly so called. On this account, many have considered these bodies useless, but this is incorrect, because they are indispensable for reproduction. This is proved by the remarkable convolutions of the nutritive vessels about them, and is confirmed by the castration of females, which is invariably accompanied by sterility. Varro writes that spayed cows will conceive, if they copulate immediately; a thing which is no doubt true of males, in which the seminal vesicles are still filled with sper- matic fluid [after castration], but not of females, in which such vesicles are not present. Hoffmann's assertion appears incorrect, a fact which we should amply demonstrate, if it had not already been done by Wharton. How the ova are fertilized and proceed to the uterus will be explained in the following chapters. REFERENCES Corner, George W.: "The discovery of the mammalian ovum" Quart. Phi Beta Pi Med. Fratern. 27:76, 1930; also in Lectures on the History of Medicine; a series . . . at the Mayo Foundation . . . 792(3-/952 (Philadelphia and London: [C 19331), 40'- De Graaf, Regner: De mulierum organis generationi insei-vientibus tractatus novus demon- strans tarn homines et animalia caetera omnia, quae vivipara dicuntur, hand minus quam ovipara ab ova originem ducere (Leyden: 1672). : "Een nie^v ontwerp van de ledematen der vrouwen tot de voort-teelinge dienstig" etc., in Alle de ivercken, so in de ontleed-kunde als andere deelen der medicyne, beschreven door de Hr. Regnerus de Graaf, en sijn leven naukerig ontleder en gelukig geneesheer tot Delft (Amsterdam: 1686). On the Female Testes or Ovaria. Translated for "The British Record" (edited by Charles Clay, M.D., Manchester) from the original edition of 16SS, by R. Knox, M.D., F.R.S.E., Edinburgh (Manchester: 1848). The copy in the Army Medical Library is boimd with: The British Record of Obstetric Medicine and Surgery, for i8.f8, vol. I. EXPLANATION OF PLATES PLATE FOURTEEN Showing the ovaries of the coav and sheep, indicating what takes place in them after coitus. Fig. 1. Showi?ig the testicles of the cow. AA. The testicle opened lengthwise. BB. The glandular substance which is found in the testes after expulsion of the ovum, cut through the yjiiddle. CC. ^ cavity, in which there was an ovum, almost obliterated. DD. Ova of various sizes contained in the ovary. EE. Blood vessels running to the ova. F. Complex membranous expansion of the Fallopian tube. G. Orifice in the end of the tubes. H. Part of the Fallopian tube cut away, with the ovary. Fig. 2. Showing the unopened testicle. A. Testicle. B. Glandular substance protruding from the testicle. C. Opening in the center of the foregoing. D. Part of the membranous expansion of the Fallopian tube. Fig. 3. Showing the testicle of a sheep with transparent ova not yet acted upon by male semen. Fig. 4^. Shoiving the glandular substance of the globules, removed from the testicle of a sheep, and still containing the ovum. A. Glandular substance of the globule, opened. B. Region from which the egg has been removed. C. The ovum, removed from foregoing. Fig. 5. Shozuing a testicle of a sheep from which the ovum zvas expelled a few days before. A. Testicle divided through the middle. B. Glandular substance of the globules with its cavity almost obliterated. CC. Ova of various sizes remaining in the superficial part of the testes. DD. Blood vessels running to the ova. E. Part of the ligament of the testicles. [132: [133] PLATE FIFTEEN Showing the testicle or ovary of the cow opened, as it is observed before coitus. AA. The testicle opened lengthiuise. B. Very large or mature ovum still contained in the testicle. CC. Smaller or immature ova remaining in the testicle. DD. Membrane of the testicle called Dartos. E. Very large ovum removed from the testicle. F. Membranous expansion of the Fallopian tube. G. Contracted orifice in the end of the tube. H. Extremity of the Fallopian tube. II. Rest of the tube. K. Detached part of the uterine horn. L. Ligament of the tube, which in humans is like a bat's wing. I 1:134] Tah XV CiSoI] PLATE SIXTEEN Showing the testicle or ovary of a woman, with the end of the tube attached. A. Testicle opened lengthwise at its lower side. BB. Ova of diverse sizes coritained in the membranous substance of the testes. CC. Blood vessels passing copiously into the center of the testis from its up- per part, and running to the ova. DD. The ligament of the testicles, by luhich they are attached to the uterus, cut. E. Detached part of the Fallopian tube. F. Cavity of the detached tube. GG. Orifice in the end of the tube. H. Foliaceous expansioyi of the tube. l.Foliaceous expansion of the tube attached to tlie testes. [136] rTah.XVI 1:137: I THE BIOLOGICAL STANDARDISATION OF THE VITAMINS By KATHAI<.INE H. COWARD From the PHARMACOLOGICAL LABORATORY OF THE COLLEGE OF THE PHARMACEUTICAL SOCIETY, UNIVERSITY OF LONDON AT THE NATIONAL INSTITUTE FOR RESEARCH IN DAIRYING SHEFFIELD, READING, BERKSHIRE ENGLAND TO Herbert McLean Evans IN WHOSE LABORATORY I FIRST HEARD OF THE EVALUATION OF BIOLOGICAL RESULTS BY STATISTICAL METHODS THE BIOLOGICAL STANDARDISATION OF THE VITAMINS SIR HENRY DALE has Said that "the object of every committee on biological standardisation is self-extermination." Chemical and physical determi- nations are more quickly carried out and, in general, are more accurate than biological determinations, and one of the chief functions of workers on biological assay is to correlate their results with those obtained by chemical and physical methods in order to find out how far these latter methods really measure the activity in question. It therefore behoves workers in the biologi- cal field not only to make their determinations as accurate as possible under the very difficult conditions inherent in the nature of their task, but also to apply to their results accepted methods of estimation of the accuracy of their determinations. Otherwise it is only too easy to blame the inaccuracy of the biological method for differences between chemical and biological results and for differences between two or more biological results. The same statistical examination of results is needed when two determina- tions appear to be equal, for there is nearly always good internal evidence from the experiment that there is a fair chance that the results are not in fact equal. The accuracy of a biological determination is generally stated as, for example, a one-in-two chance (50:50) that the result obtained lies outside, plus or minus a certain percentage, of the true result ("the probable error"), a one-in-three chance that it lies outside somewhat larger limits ("the standard error"), a one-in-twenty-two chance that it lies outside larger limits, a one-in-a-hundred chance that it lies outside larger limits still. These chances are often stated as probabilities that the result lies within the corresponding limits when they are designated as the limits of error for P = 0.50, 0.67, 0.95, 0.99 respectively. Every biological determination should be stated as the result obtained to- gether with the limits of error corresponding to the probability most likely to be of use in evaluating the result. The Sub-Committee on the Accuracy of Biological Determinations, of the British Pharmacopoeia Commission, chose to express the limits of error of the tests it described for P = 0.95 as being the figure in rather general use, and also for P = 0.99 as being a figure which a court of law would probably consider large enough in possible legal proceedings.^ Essentials of the Biological Method Workers on the biological assay of vitamins now generally recognise certain essentials for their task. Some of these were apparent almost at the beginning of the attempts to measure these factors; others have become evident only through arduous and most painstaking labours. 1. Standard of Reference— Every worker is familiar with the variation in 1 44 Standardisation of the Vitamins Ik response of different animals to the same dose of any vitamin. Every worker who has tested the same dose of a vitamin at several different times is familiar with the fluctuation in the average response of the groups of animals used, and hence he has realised that it is impossible to make a direct comparison be- tween a result obtained with a dose of a particular preparation at one time with the result obtained with a similar dose of another preparation at a different time. For example, it is impossible to compare summer and winter milk by determining the average responses of two groups of rats which were dosed in the summer and winter months respectively. The only way of com- paring summer and winter milk is by comparing each separately with a prepa- ration known to be stable throughout the interval elapsing between the tests and thus indirectly to make the comparison between the two samples. The same method is even more necessary if results obtained in one laboratory are to be compared with results obtained in another. A standard of reference for this purpose may be any preparation of the sub- stance to be determined provided it conforms to accepted criteria of purity; but if there is any doubt as to the possibility of obtaining the substance in a nearly enough pure form, or, as has often happened, a standard is required before even the identity of the substance has been recognised, then one preparation of the substance in possibly a very impure form should be made available to workers throughout the world and a particular weight of that particular prep- aration should be accepted as a unit of the activity it is designed to measure. When the active substance itself becomes available in acceptable form, then the impure standard and its unit can be discarded. Probably by that time, the chemistry and physics of the vitamin will be so far known that the potency of a preparation for therapeutic use will be determinable by chemical or physical methods, although the activity of a food substance may, through difficulties of extracting the vitamin quantitatively, still have to be determined by a bio- logical comparison with the pure substance or the original standard of refer- ence. If compared with a sample of the pure substance which is not the actual preparation adopted as an International Standard, then the potency of the food substance should be stated as the weight of active substance per gram, not as a number of International Units. Up to the present, four standards of reference for vitamin determinations have been adopted for international use by the Health Organisation of the League of Nations.^ For vitamin A, in 1931, a preparation of carotene was made available. With increased knowledge of the chemistry of carotene, this preparation proved to be a mixture of a- and ^-carotenes, and in 1934 it was replaced by a pure sample of ^-carotene, the weight of the new preparation accepted as a unit being adjusted so that the unit of activity remained the same. For vitamin B,^ a preparation made by Prof. B. C. P. Jansen was adopted which was an adsorption product on fuller's earth from an extract of rice polishings. When the supply was nearly exhausted a similar preparation was Katharine H. Coward i^r made by Professor Jansen and biological tests carried out by workers in several laboratories showed that its potency was equal to that of the original prepara- tion. W^hen vitamin B^ was synthesised, a large sample of it was adopted as the International Standard of reference and workers in eleven different labo- ratories established the relation between the pure substance and the original standards. Three micrograms of the crystalline material were equivalent in biological activity to lo mg. of the standard, that is, to i International Unit.' For vitamin C, in 1931, there -was no known preparation stable enough to use as a standard of reference, and the juice of fresh lemons, thought from its antiscorbutic effect on guinea-pigs not to vary greatly in potency, was adopted. By 1934, however, ascorbic acid had been prepared in a pure state and Prof. A. Szent-Gyorgyi prepared 200 gm. for use as an International Standard of reference. Since then ascorbic acid has been synthesised and the British Phar- macopoeia now accepts chemical and physical tests of purity without biolog- ical evidence of activity.* For vitamin D the original preparation of irradiated ergosterol is still in use. For vitamin E a standard of reference will almost cer- tainly have been adopted by the time this book is in print. It is a preparation f//-a-tocopherol acetate of well-defined chemical and physical properties, shewn to be stable for many months, not only when kept under the best conditions known but also during the varied conditions of actual dosing in the experi- ments carried out by workers in twelve laboratories to determine its activity and the shape of the curve of response given by female rats. Other standards of reference will become necessary as further vitamins are discovered (for example, vitamin K), but it is possible that their chemistry and physics may be so well known by the time their function is recognised that therapeutic prepa- rations, at least, may be determined without biological tests. 2. Criterion for the Test.— hong before any attempt is made to determine the amount of a vitamin in any substance, a fair idea will have been gained as to its action on some animal suitable for laboratory experiments. Sometimes a vitamin has several functions; for instance, vitamin B^ is necessary for growth (increase in weight in rats), for preventing or curing polyneuritis in pigeons, and for preventing or curing "convulsions" in rats. When increase of weight is the criterion adopted, it is specially important that the basal diet should contain abundance of all substances necessary for growth other than the vita- min in question. This criterion for vitamin B^ has lower limits of error (79- 126 per cent for P = 0.95) than the cure of polyneuritis in pigeons Avhose limits of error are 24-417 for P = 0.95. The cure of convulsions in rats produced by a deficiency of vitamin B, and cured by giving that substance has a high degree of accuracy in the hands of Dr. E. M. Nelson.' It must never be assumed that any method has the same degree of accuracy when performed by all workers, nor indeed by any one worker in all his tests.' All criteria but increase in weight have the disadvantage of uncertain start- ing-points'and end-points. The onset of xerophthalmia is ill-defined, as also 1 46 Standardisation o£ the Vitamins its cure. The length of time taken to effect a cure is therefore difficult to estimate. Even the increase in weight is not very accurate, for a rat's weight varies almost from hour to hour and may not be the same (or increase steadily) at the same time on successive days. 3. Adequacy of the Basal Diet.— ¥or a simple comparison between two sub- stances (for example, the standard of reference and a food substance) with respect to a particular vitamin, it is obvious that the diet of the animal used should contain abundance of all substances known to be necessary for the well-being of the animal, except the one vitamin which is to be assayed. Otherwise, the food substance or even the standard preparation may supply some factor which is helpful to the work of the vitamin under assay and the restdts may thereby be vitiated. A slight deficiency of one of the vitamins does in fact limit the steepness of the curve of response to another vitamin.' 4. Statistical Examination of Results.— More and more workers are submit- ting the results of their experiments to statistical examination to determine how much confidence may be placed in them. The accuracy of an assay de- pends on three factors: (1) the standard deviation of the individual observa- tions, (2) the number of animals from which the average result is calculated, and (3) the steepness of the curve relating response to dose of vitamin given. It is evident that the less the individual results differ from each other, the greater confidence one may have in the reliability of the average, and that the larger the number of animals used, the more nearly will the average approach the true value. It is only necessary to draw two curves of response, one much steeper than the other, to discover that a particular standard deviation of a result corresponds to a smaller deviation of dose in a steeper curve than in a less steep curve. The significance of a difference between two means is deter- fji j^i mined by a simple formula /= — ^ " and by reference to a table of t V^i' + ea^ which indicates the probability of the difference being significant.* When ^=3 and the number of animals used has been about twenty, it is generaly consid- ered that the difference found is significant. Some workers of a more optimistic temperament are satisfied with a value of 2 for t. It can only be suggested that these should repeat their experiment on the same or similar lines. The selection of doses for an assay leaves something to choice. If any dose produces a maximum or nearly maximum result, or a minimum or nearly minimum result, it cannot be used in the calculation of potency. Some workers like each experiment to be self-contained and to give one group of rats a dose of the standard and several different groups of rats different doses of the sub- stance under test. They then equate the dose of standard to the dose of test substance which gave the result most nearly equal to that given by the dose of standard. This way of calculating the result makes no use of the informa- tion furnished by all the rats in the other groups which were given other doses. The defect can be remedied by drawing a smooth line as nearly as Katharine H. Coward i^h possible through the points plotted from the results of the different doses and then finding the dose corresponding to the result obtained from the one dose of standard. If the relation between effect and dose is logarithmic, the line relating effect to the logarithm of the given dose is a straight one and can be calculated mathematically with very little difficulty. The whole process is then independent of judging where exactly the line should go. Other workers like to give two groups of rats two different doses of the standard and two other groups two different doses of the test substance. Then two curves of response are obtained, their average slope determined and the relative potencies of test and standard found by the distance between the tAvo parallel lines. On the other hand, some -workers prefer to establish a general curve of response for their test and compare the results from dose of standard and dose of test substance through the doses corresponding to these results on the general curve. Unless it has been shown that the slopes obtained in various experim.ents are significantly different from each other, then the method of using the general curve of response is probably tlie best one, as having been determined by the use of a much larger number of rats than would be used in any one assay alone. It is well worth while to realise, as R. A. Fisher has so clearly shown in his Design of Experiments,^ that, by careful planning, an experiment can be made to yield far more information than the one point primarily needed and that by such planning the information on the chief point is also made much more reliable. For example, the writer has lately compared two sam- ples of irradiated ergosterol for vitamin D activity. It was also desired to discover without independent experimentation whether the differences in response to vitamin D were greater between different litters than between different rats of the same litter. Several years ago the writer had sliown that tliere was a fair correlation between the amount of liealing brought about by vitamin D and the weight of the rat when it was first given the rachito- genic diet. By careful arrangement of the rats within each litter given the particular doses (three of one sample of irradiated ergosterol and three of the other sample) it -was possible to arrange the results {a) as a "scjuare" in which the row's were litters and the columns were doses, and {b) a second "square" in which the rows were litters and the columns were rats in ascending order of initial weight. Thus, information was gained on two points of impor- tance in addition to the one for which the experiment was originally needed. Discarding the principle of standardising the conditions of experiments by varying only one factor at a time, but varying several conditions instead, makes it possible to gain not only more information on different points but also information on the interaction of these conditions. The interaction of the vitamins, already indicated almost by chance, is only one of many fascinating lines which this method of procedure suggests in planning c\en routine assays. 148 Standardisation o£ the Vitamins REFERENCES 1. British Pharmacopoeia Commission: Report of the Sub-Committee on the Accuracy of Bio- logical Assays (London: 1936). 2. Leag. Nats., Quart. Bull. Hlth. Organisatn. 3:428, 1934. 3. Leag. Nats., Bull. Hlth. Organisatn. 9:371, 1940-41. 4. British Pharmacopoeia: Addendum ip^6 (London: 1937), 4. 5. Kline, O. L.; Tolle, C. D., and Nelson, E. M.: Jl. Assn. Offic. Agricult. Chemsts. 21:305, 1938. 6. Coward, K. H.: The Biological Standardisation of the Vitamins (London: 1938), 196. 7. Coward, K. H.: The Biological Standardisation of the Vitamins (London: 1938), 138. 8. Fisher, R. A.: Statistical Methods for Research Workers (7th ed.; Edinburgh, London: 1938). 177- 9. Fisher, R. A.: The Design of Experiment (2d ed.; Edinburgh, London: 1937), 118. THE PREVENTION OF DEAFNESS By S. J. CROWE X From the OTOLOGICAL RESEARCH LABORATORY OF THE JOHNS HOPKINS UNIVERSITY BALTIMORE, MARYLAND I T THE PREVENTION OF DEAFNESS HE WHITE HOUSE CONFERENCE Called by President Hoover in 1931, and composed of public health officials, social workers, psychiatrists, those in- terested in child labor, and members of other agencies dealing with the welfare of children, estimated that three million school children in the United States are handicapped by impaired hearing. This figure probably represents only the more advanced cases. Our impression after seventeen years of study of the causes and prevention of impaired hearing is that deafness is incipient in a much greater number, and in many it will slowly progress and become a handi- cap later in life unless something is done to correct the cause in the early stages of the disease. Almost every variety of deafness in adults is also found in children. Many types, such as otosclerosis, hereditary nerve deafness, Meniere's disease, acoustic tumors, and developmental anomalies, cannot be helped by operation or treatment, but in the large group caused by infections and me- chanical interference with the function of the Eustachian tubes the progress of the disease can be stopped and often the hearing can be restored, provided the condition is recognized early and treatment instituted before the middle- and inner-ear structures are damaged. After years of trial of the method used in general pathology, namely, the correlation of clinical examinations and functional tests with the gross and histologic appearance of the middle and inner ear after death, my associates and I have become convinced that most of the remedial kinds of deafness begin in childhood, and that to learn more about causes, treatment, and prevention we must concentrate our studies on living, grooving children. Deafness in adults is rarely cured, but it is truly astounding hoAV often hearing can be restored in children under fourteen with a middle-ear type of deafness severe enough to place them in a lip-reading class. With the cooperation of the Baltimore Department of Health and Board of Education we have examined for three successive years the hearing and upper air passages of an unselected group of 1,365 school children between the ages of eight and fourteen years. The hearing of each child has now been tested from eight to ten times in sound-proof rooms, for all octaves from' 32 to 16,384 d.v. We were fortunate to have for this purpose audiometers built by the Western Electric Company for research. Commercial audiometers are not designed to test the frequencies above 8,000 d.v. and for this reason the fact had not been previously observed that almost every variety of middle- as well as inner-ear deafness begins with impaired hearing for the higher tones, and progressi\'ely involves the lower tones and the speech range (250 to 3,500 d.v.). In addition to the hearing tests, the tympanic membranes, nasal passages, nasopharynx, and particularly the pharyngeal orifice of the Eustachian tubes were examined with a nasopharyngoscope, which provides as clear a vicAV of these regions as a cystoscope does of the interior of the bladder. All but a small percentage of these 1.365 children were thought by their teachers and 152 Prevention o£ Deafness parents to have good hearing. We found, however, that only 58.8 per cent have normal hearing for all tones; 36.3 per cent have impaired hearing for the three or four octaves at the upper end of the scale, with normal hearing for all lower tones and for the voice— in quiet surroundings. This picture- impaired hearing for high tones with good hearing for low tones and for the voice in a quiet place— is identical with the common type of partial deafness in old age. Elderly people may understand conversation perfectly if the fre- quencies used in speech (250 to 3,500 d.v.) are not masked by other voices or by the roar of traffic. Under the latter conditions, however, they have difficulty in understanding unless the speaker raises his voice above the decibel level of the surrounding sounds. All of us with normal hearing throughout the entire scale would be similarly handicapped in noisy surroundings if we had not unconsciously trained ourselves since childhood to interpret properly the overtones of words, just as a stenographer in reading her notes interprets shorthand symbols. Most of our city schools are on streets with heavy traffic, and children with impaired hearing for only the higher frequencies have diffi- culty in hearing the teacher when there is a background of noise in the class- room. This leads in many instances to an erroneous diagnosis by the teacher of indifference, stupidity, or mischievousness, and explains their failure to do well in their school work. Lymphoid hyperplasia in the pharynx and nasopharynx is responsible for ear infections, and through interference with the ventilating function of the Eustachian tubes is an important cause of impaired hearing in childhood. Thousands of children would be spared the disabilities of impaired hearing and other complications of enlarged and infected lymphoid tissue, if we un- derstood better the basic problem of why some children are more susceptible than others to respiratory infections, and knew more about the relation be- tween allergy, dietary deficiencies, and ductless gland disorders to lymphoid hyperplasia. The cause could then be corrected at the source, and it would rarely be necessary to remove tonsils and adenoids or operate for infected ears. A large percentage of childhood deafness is caused directly or indirectly by hypertrophy and infection of lymphoid tissue in the nasopharynx. Infec- tions are easily recognized and as a rule are adequately treated with chemo- therapy and operation. The greatest danger to the hearing during childhood is partial or intermittent closure of the Eustachian tubes by hypertrophied nod- ules of lymphoid tissue in and around their pharyngeal orifice, and the thick mucous secretion that accumulates in the tubes, the middle ear, and in the pneumatized spaces throughout the temporal bones. These changes in the middle ear and tubes are not associated with pain, sudden loss of hearing or other symptoms that would direct the attention of the child, the parents, or the family physician to the ears. The earliest evidences are retraction and opacity of the tympanic membranes and impaired hearing for the tones above 8,000 d.v. The hearing defect at this stage can be demonstrated only with the audiometer. Impaired hearing for the high tones is also due to causes that S. J. Crowe i^o cannot be cured, but this finding in a child demands an examination of the nasal passages and nasopharynx with a nasopharyngoscope, and a correction of any condition that interferes with Eustachian-tube function and so leads to a further loss of hearing. Lymphoid tissue is an integral part of the mucous mem- brane of the pharynx and nasopharynx and cannot be removed unless the entire thickness of mucous membrane is taken out. After removal of tonsils and adenoids before the age of puberty, it is a common occurrence to see nu- merous nodules on the posterior and lateral walls of the pharynx— a "graiadar pharyngitis." A similar condition is almost invariably seen in the nasopharynx if the examination is made with a nasopharyngoscope. These nodules may not be large enough to cause mouth breathing, but nevertheless are a source of potential danger to the hearing. Physicians are too prone to think that ade- noids can be removed in their entirety, and after operation the nasopharynx is dismissed withotit further examination as a source of trouble. Recurrence is so frequent, however, that it must be regarded as normal. It is not the size but the location of lymphoid nodules that leads to deafness. Frequently children with beginning deafness due to these causes sponta- neously recover their hearing, or further loss is prevented by the regression of lymphoid tissue and restoration of the patency of the Etistachian tubes which occurs at puberty. Our duty is to recognize the fact that good hearing depends on open Eustachian tubes, to examine school children with this point uppermost in our mind and to institute some form of safe, conservative therapy that will tide the child over the dangerous period, which tisually ends at puberty. Every year the hearing of millions of children is tested with some type of audiometer. Those with severe impairment are recognized and placed in vocational classes, but the incipient cases, the ones which can be benefited the most, are either not recognized or are inadequately treated. Surgical re- moval of tonsils and adenoids often fails to remedy permanently the conditions that may cause deafness; certainly a second or third operation is not the proper treatment for a recurrence. It has been known for nearly forty years that lymphoid tissue is more sensi- tive to irradiation with X ray or radium than any other structure in the body. Therefore in treating hyperplastic lymphoid tissue with these rays the size of the mass may be reduced in three or four treatments with a dosage so small that there is no danger of a burn, or injury to bone, cartilage, hypophysis, or central nervous system structures. The life cycle of the lymphocyte is short, certainly less than a month. Irradiation treatment effects a reduction in size of a mass of lymphoid tisstie by retarding or entirely stopping cell division and the formation of new lymphocytes to replace the ones that have been cast off. For nearly twenty years we have used X rays and radon, sometimes as a substitute for, and often to supplement, surgical removal of tonsils and adenoids in children, and have come to the conclusion that radon is a much simpler, safer, and more effective method for irradiation of the nasopharynx. The objections to the use of X rays for this purpose are: the difficulty of direct- 154 Prevention o£ Deafness ing the central beam of the rays from the various portals in the neck so that they will cross or converge at the exact spot where they are needed; the time consumed and the expense of the treatments, which must be divided over a period of two or three weeks; a large percentage of the rays is absorbed by skin and muscle, and never reaches the nasopharynx; therefore the total irradiation used is much greater than the amount necessary to reduce the size of adenoids. It is not wise unnecessarily to irradiate the base of the skull and centers of ossification to this extent in a growing child, especially since the series of treatments may have to be repeated several times to maintain the patency of the tubes. The chief objection is the inaccuracy of this method in young children. On the other hand, radon can be compressed and an amount equivalent to a gram of radium salts may be sealed in a capillary glass bulb, placed in an applicator small enough to pass along the floor of the nose without anaesthesia and without pain, and brought into direct contact with the tissue to be treated. The applicator remains in place from three to five minutes, depending on the number of millicuries it contains. The effect of the treatments is determined by direct inspection with a nasopharyngoscope and by audiometer tests. When necessary the treatments are repeated at inter- vals of a month, but rarely more than three or four are required. We have often irradiated from fifteen to tw^enty school children in an afternoon. Many children have been followed for a decade or more and we are confident that no untoward local or systemic effects occur. Partially occluded Eustachian tubes set in train a series of physical changes in the membranes of the middle ear and mastoid cells that eventually impede the movements of the ossicles. The earlier the condition is recognized and treated the better the results. For this reason children over fifteen and adults are rarely benefited. Irradiation does not cure any and every type of deafness. Its purpose is to remove lymphoid tissue from locations that will cause deafness. Detailed results of a study of 1,365 school children have been published and need not be recounted here. It is a type of experiment that requires years for its com- pletion, and further reports will appear from time to time. The method is safe and inexpensive and if applied in schools would in time greatly reduce the recent estimate of three million children with seriously impaired hearing. S. J. Crowe i^^ Relevant Literature 1938 Crowe, S. J., and Guild, S. R.: "Impaired hearing for high tones" Act. Otolaryngol. 20:138, 1938- Crowe, S. J.: "Diagnosis and differential diagnosis of deafness" Archs. Otolaryngol. 28:663, 1938- 1939 Crowe, S. J., and Baylor, J. W.: "The prevention of deafness" Jl. Amcr. Med. Assn. 112:585, 1939- Crowe, S. J.: "The prevention of deafness in children" Laryngoscope 49:591, 1939; also in Trns. Amer. Otol. Soc. 29:112, 1939. 1940 Crowe, S. J.: "The recognition, treatment and prevention of hearing impairment in children" Laiyngoscope 50:658, 1940; also in Trns. Amer. Otol. Soc. 30:27, 1940. Burnam, C. F.: "Irradiation treatment of hyperplastic lymphoid tissue" Laryngo.scope 50:663, 1940; also in Trns. Amer. Otol. Soc. 30:32, 1940. Polvogt, L., and Babb, D. C: "Histologic studies of the Eustachian tube of individuals with good hearing" Laryngoscope 50:671, 1940; also in Trns. Amer. Otol. Soc. 30:41, 1940. Guild, S. R.; Polvogt, L. M.; Sandstead, H. R.; Loch, W. E.; Langer, E.; Robbins, M. H., and Parr, W. A.: "Impaired hearing in school children" Laryngoscope 50:731, 1940; also in Trns. Amer. Otol. Soc. 30:47, 1940. Bordley, J. E.: "The treatment and prevention of deafness in children" South. Med. Jl. 33: 1159.1940. Crowe, S. J.: "The nasopharynx" Trns. Amer. Laryngol. Assn. 1940:145; also in Archs. Oto- laryngol. 33:618, 1941. 1941 Crowe, S. J., and Burnam, C. F.: "Recognition, treatment and prevention of hearing impair- ment in children" Ann. Otol., Rhinol. & Laryngol. 50: 15, 1941. 1942 Crowe, S. J.; Guild, S. R.; Langer, E.; Loch, W. E., and Robbins, M. H.: "Impaired hearing in school children." Laryngoscope 52:790, 1942. GENE H AND TESTOSTERONE IN THE FOWL By C. H. DANFORTH From the DEPARTMENT OF ANATOMY STANFORD UNIVERSITY, CALIFORNIA 1 GENE H AND TESTOSTERONE IN THE FOWL IN MANY birds the plumage of the male is appreciably different from that of the female. Such a difference is particularly noticeable in the Phasianidae , of which the common fowl is a familiar example. But there are also species in which this dimorphism in plumage does not occur, and even in the domestic fowl some races fail to show it. In these races, while other secondary sexual characters are essentially as usual, the plumage of the male is almost identical with that of the female. This type of feathering is called henny* and has been shown to be clue to a dominant autosomal gene designated as H. Males that are homozygous {HH) and usually those that are heterozygous (H/i) for this gene are characterized by henny plumage, while those homozygous for the recessive allele {hh) develop the normal cocky plumage. Females of the formulas HH and hh are indistinguishable except by genetic or involved endocrine tests. This interesting situation raises the question of how gene H is related to other genes and how it affects feather development. In this paper the role of gene H in relation to testosterone will be considered. The heredity of hen-feathering in the male was studied many years ago by T. H. Morgan, by R. C. Punnett, and by a number of others, all of whom were fairly unanimous in their interpretation of the genetic behavior of the trait. More recently Punnett^ has suggested that the normal locus of the H allele is in the Y chromosome of the female and that it becomes autosomal and po- tentially present in the male only through translocation. This hypothesis may explain several puzzling observations that are on record, but whatever the exact cytological relations may be, the somatic manifestation of the gene has been known for a long time and in many different breeds. There are several records of new appearancesf which may reasonably be attributed to fresh mutations or, on Punnett's theory, to translocations following earlier muta- tions. In any event, when the gene once appears in an autosome its subsequent genetic behavior is fairly simple. There is no evidence known to the writer which suggests that the trait may have a prototype in any wild species. * In early days when cock fighting was permissive in England there were some pit games in which the males Avere hen-feathered and, because of the supposed advantages accruing from such a camouflage, hen-feathered cocks attained a certain popularity. In the argot of the day they ^\•ere called "hennies" — whence, presumably, our current scientific term henny and, by analogv, its correlative cocky, terms which are now used to characterize the two dis- tinctive types of phuiiage that may occur in male fowls. f As long ago as 1873, Tegetmeier," whose name appears often in Darwin's Animals and Plants under Domestication, recorded one fresh occurrence of the trait. A game bantam "of known purity" which hatched in the spring of 1859 at first developed normal plumage, after which it became 'strictly hen-feathered both as to form and colour" and produced some hen- feathered male offspring. Others of its offspring were normal and took second prize at the Crystal Palace show. It was exhibited at a meeting of the Zoological Society on March 16, 1861 . "Nothino would have been easier than to have established a permanent breed," said Teget- meier — and he was probably right. [ '59;] 1 6o Gene H and Testosterone in the Fowl The experimental form used for this study was the Sebright bantam. This is a small breed established about the year 1800 by Sir John Sebright, who combined the color of a crested Polish, the size of a bantam, and the henny plumage of another fowl in the "laced bantam" which later came to bear his name.* The breed in its "golden" and "silver" phases was apparently soon perfected and has remained practically unchanged for many years. Colored illustrations done by Harrison Weir about 1870 show Sebrights that would be favorably rated today. Because of its distinctive traits, stabilized genetics, and small size, the Sebright has become a favorite with experimenters. A certain amount of lore concerning the control of differences between male and female plumage dates back to the nineteenth century, but it was not until well into the present century that the subject became a matter for active, critical experiment and analysis. Work in this field is well summarized by Pezard, Sand and Caridroit,° by Greenwood and Blyth," and by Domm.' Briefly stated, it has been found that in the brown Leghorn and other hh breeds presence of the testes or injected male hormones has little influence on the plumage, but functional ovaries or circulating estrogens have a feminiz- ing effect on developing feathers of either sex. These findings, which have been verified on many breeds, might have been considered of general applica- tion to all fowls had it not been for the atypical reaction characteristic of the Sebright. Soon after Goodale^ had demonstrated the effect of ovariectomy in the brown Leghorn it occurred to Professor T. H. Morgan that the Sebright testis might produce a certain ainount of female hormone and thus account for the henny plumage. His experiments"' ^"'^ showed that removal of the testis did in fact lead to the production of cocky plumage in the hitherto hen-feathered male. So-called luteal cells were tentatively identified in the Sebright testis and for a time it seemed that the explanation had been found for henny plumage, not only in the Sebright bantam but in wild species in which the sexes are similar. This view was not long sustained. Nonidez^" showed that the "luteal cells" were not gland cells; Roxas" demonstrated by exchange * The account of the synthesis of this breed is repeated in many of the older books on poultry. Sir Thomas Sebright's report, which is quoted by Dr. Horner in Tegetmeier"s volume, is apparently authentic and also cjuite plausible genetically. The statement in the Encyclo- pedia Americana^ (1941 ed., vol. 3, p. 215), that "the golden and silver Sebright bantams origi- nated in America," is apparently entirely withoiu foundation. We may also discount the interesting notion of the Rev. Mr. Dixon^ (1848) who linked in some mysterious way the sprightly virtues of the diminutive fowls with those of the dainty human inhabitants of Bantam, Java. Some French writers have implied that bantams originated in England, but the "grigs" of seventeenth century England were not the original dwarf fowls. Plin) is one early writer who makes no attempt to trace them to some remote source, contenting himself with: "Est et pumilionum genus non sterile in his, quod non in alio genere alitum, sed quibus certa fecunditas rara, et incubatio ovis noxia." {Nat. Hist. X:77.) Probably dwarf specimens have appeared from time to time in many large breeds, but with genes for nannism once in their armamentarium it has been easy for breeders to produce bantam forms at will. In the 1941 exhibit at the California State Fair a trio of "Black Minorca Bantams" deemed worthy of a premium were said by their exhibitor to have had a full-sized black Minorca grandparent. I C. H. Danforth 261 operations that in their effects on plumage Leghorn and Sebright testes are interchangeable; and Danforth and Foster'^ found by transplantation of skin that the differential factors involved actually reside in the skin itself, presum- ably in the feather follicles. Two pieces of skin healed together and growing adjacent to each other on the same host will, if one is hh the other HH, pro- duce feathers respectively of the cocky and henny type. This observation was substantiated by Champy'^ who transplanted skin of a Phoenix cock to a Sebright and got typical male plumes surrounded by the henny feathers of the host. Nevertheless, as late as 1936, when Callow and Parkes" failed to get a response in the plumage of Sebright capons injected with androsterone, it was again suggested that henny plumage of the Sebright cock must be due to another substance secreted by the testis, "probably oestrin." Later experi- ments with other androgens have caused this opinion to be revised. Gallagher, Domm and Koch,^ and Danforth" next showed that extracts from bull testes would "feminize" the plumage of Sebright capons, and the latter also found that an ordinarily subthreshold dose of the extract could be made to supplement a hitherto inadequate output of a testicular fragment. Perhaps even more interesting was the demonstration, which has been re- peated in connection with the present experiments, that the comb, "psyche," and feather follicles have different threshold levels, the level for the latter being appreciably higher. Taking advantage of this fact one can, by incom- plete castration or injection of sufficiently small amounts of testosterone into capons, convert genetically HH birds into phenocopies of the hh form that would be likely to deceive even a critical observer. Transplantation of avian testes, and injection of extracts from testes of mammals, is not entirely satisfactory, however, since the composition of the extracted hormones is at least somewhat uncertain and there is also the possi- bility that the effects observed may be mediated indirectly through some other gland. The latter possibility is mentioned again and again in the litera- ture, particularly with reference to the thyroid. Since the discovery by Torrey and Horning^'-' of the effect of desiccated thyroid on plumage the question has been discussed, pro and con, by Crew;^ Danforth;"^ Parkes and Selye;"' Emmens and Parkes;"" and Chu."" The relation of hypophysial and suprarenal secretions to plumage types is still less well known. Various aspects of their possible relations have been considered by Danforth;'"'^ Hill and Parkes;"® and Witschi.^ A survey of this literature, which need not be undertaken here, makes it apparent that with the partial exception of those on the thyroid and ovary the experimental studies thus far have not sufficed to differentiate be- tween the direct and indirect effects of the rather numerous hormones which show evidence of playing some role in the determination of plumage character- istics. In an attempt to meet this objection in the case of testosterone the synthetic drug* was administered in various ways to Sebright capons. * The writer is indebted to the Sthering and Ciba firms for generous supplies of testosterone propionate made up in tire desired vehicles. 1 62 Gene H and Testosterone in the Fowl Synthetic testosterone propionate dissolved in sesame oil was injected in 0.5—1.0 mg. doses into the breast muscles of young bantam capons which had already developed cocky plumage. In HH birds the usual effects were ob- tained: a slight temporary drop in weight or retardation in growth, an almost immediate increase in size and color of the comb, an increase in activity and morale (which, however, is by no means entirely under humoral control), and a more or less complete "feminization" of such potentially dimorphic feathers as happened to be growing at the time of injection. Similar results with syn- thetic testosterone have been obtained by Parkes and Selye,"* Champy,^" and others. In this series there was some evidence, difficult to measure, that the testosterone slightly accelerated growth, which perhaps helps to account for a somewhat greater than usual length of some of the feminized feathers (pi. 1, fig. 1, d). The similarity of this testosterone effect to that produced by theelin is indicated in plate 1, figure 2, where there are shown partially and completely "reversed" feathers resulting from both types of treatment. The foregoing experiments throw no additional light on whether the action of the hormone is direct or indirect. This phase of the problem was next at- tacked by the use of relatively minute quantities of testosterone which was tested for its direct local action— a technique which has been used successfully in studying the effects of estrone by Lyons and Sako''^ with mammals, and Greenwood and Blyth^" with birds. For these purposes testosterone propionate was employed as pure crystals, as an ingredient of a salve, and as an ointment dissolved in sesame oil. The oil solution was applied directly to the skin, beginning soon after an area on the back had been plucked. Usually about thirty feathers were re- moved from each area to be tested, with others from one or more control areas at a distance. Plate 1, figure 1, shows the effect of such treatment. Similar results were obtained with both the silver and the golden Sebright. The chief objec- tion to this method is that the oil tends to diffuse widely over the surface and cannot easily be confined to a limited area. It is sufficiently selective, however, to reveal that regenerating feathers in the treated area are more affected than those at a greater distance. The salve proved more satisfactory for topical application than did the oil. When applied over any appreciable area the observed effects were similar to those following the application of the oil solution and generally involved slight, but definite, effects on the comb as well as the feathers. Better results were obtained from applying it in very small amounts to individual pin feathers. The procedure was to find an area in which at least one pinfeather was barely visible above the skin and then pluck all the surrounding feathers and pinfeathers. The selected pinfeather was then painted with a tiny bit of salve at regular intervals. A protocol will serve to illustrate the method and results. Golden Sebright, hatched Apr. 20, castrated May 27. Aug. 9: weight 440 gm.; comb 35 (length 25 mm. + width 7 mm. + height 3 mm.), pale. An area about 25 mm. in diameter plucked clean C. H. Danforth 163 except for three pinfeathers, the middle of which was selected for treatment. Approximately 10 mg. of salve, presumed to contain 0.02 mg. of testosterone propionate applied with the tip of a small camel's-hair brush. This treatment was repeated daily from Aug. 11-22 inclusive. Aug. 22: weight 495 gm.; comb 38, pale. Aug. 29: weight 525 gm.; comb 36. When fully developed the treated feather and one other of the same age showed a cocky tip and henny vane, resembling somewhat the middle specimens in figure 2. The third feather of the same age was somewhat intermediate, but at least a dozen younger ones were almost completely feminized and similar to the two on the right in figure 2. Others, medial and slightly anterior, were intermediate, while still others further anterior were cocky and ap- parently entirely uninfluenced. It is doubtful if the comb was really affected at all in this case in spite of the rather widespread effect on the feathers. Considering the number of feathers influenced, it would appear that when applied to the skin as a salve less than 1 .5 gammas of testosterone propionate a day for each feather is adequate to produce feminization of the plumage with little, if any, other effect. The crystalline testosterone was applied by pushing small particles into the skin. The exact amount administered was generally difficult to determine since the pieces frequently crumbled somewhat in the process of being embedded. Similar pieces, or group of pieces, carefully compared with the original were weighed on a delicate balance and the amount implanted estimated on that basis. Another protocol will show the procedure and results with this method. Golden Sebright, hatched Apr. 20, castrated May 27. July 30: not weighed; comb about 31. Seven cm. anterior to the uropygial gland, areas 23-24 mm. in diameter were plucked on each side of the midline. In the center of the right area a bit of testosterone propionate, not more than 0.5 mg. in weight, was pushed into a pocket in the skin and the whole area covered with a protecting film. The left side was treated in exactly the same \\ay except that no testosterone was administered. Aug. 5: weight 450 gm.; comb 37, tip bleeding. Aug. 9: weight 470 gm.; comb 41, red. Aug. 15: weight 480 gm.; comb 39. Growth of pinfeathers greater on the right side. Aug. 29: Weight 590 gm.; comb 40, fairly good color. New feathers on left all definitely cocky in form and color. Those on the right similar except for three in the center of the area. Two of these appear to be typically henny, one intermediate. When fully groAvn the two former proved to be entirely henny in form and color, but rather longer than usual. Figure 3 shows similar results in an experiment in which a silver Sebright was used. It would appear that in these cases enough testosterone -svas absorbed to raise the general level abo\e the threshold for comb response but not to the threshold for feather response, except in the im- mediate vicinity of the implant itself. These experiments seem to make it clear that the effects observed in HH birds following the injection of testosterone are not due to any general effect mediated indirectly through the thyroid. They do not eliminate the possibility that thyroid secretion and possibly several other hormones are ordinarily more or less involved in the production of henny plumage. At the present time we know that in birds of the formula HH which have been deprived of their testes but not their thyroids the reaction of developing pinfeathers can be changed from that shown in plate 1, figure 1, ^, to that in plate i, figure 1, c, by the application of thyroxin, of estrone, or of testosterone, and that the effects of the two latter are direct and local. Whether or not there are other hormones 1 64 Gene H and Testosterone in the Fowl which separately and independently will give the same effect remains to be determined. Capons of the formula hh respond in the same way to thyroxin and estrone but not to testosterone, and this suggests once again that it is not so much the hormone as the way in which the tissues have come to respond to it that is of fundamental importance. In the case under consideration one simple genetic alteration— a change in the character or position of a single gene— suffices to convert tissues of the feather follicle from a condition of sensitivity to estrone and indifference to testosterone to one in which they are apparently insensible to the difference between these two hormones and respond equally to both of them. In this paper the convention has been followed of considering the deposi- tion of pigment as if it were a passive phenomenon, but of late it has been shown that the pigmentophores have a great deal of autonomy and may play a definite and largely independent role in the determination of pattern (Willier and Rawles;^^ Danforth^*). Studies on plumage have paid little attention to the question of whether or not gene H affects the pigmentoblasts directly, or only alters the character of the "parade ground" on which they may deploy. It may be remarked in passing that in plate 1, figure 3, here, and in many figures by Parkes and others, both the laced henny and the striped cocky feathers are shown to be characterized by the alignment of black pigmentophores along the most distal barbules in the vane wherever those barbules may be. This sug- gests that the underlying factor may be related more to pigmentophore orienta- tion than to mere pigment deposition, but this is a problem yet to be solved. In conclusion it may be pointed out that thus far we seem to have gotten no inkling as to the underlying basis for sexual differences in plumage, but we have succeeded in gathering considerable information about some of the factors which condition the appearance of one or the other of the alternative types. That the whole situation may be fundamentally altered by a simple cytological change is apparent in the effects produced by gene H. REFERENCES 1. Punnett, R. C: Jl. Genet. 35:129, 1937. 2. Tegetmeier, W. B.: The Poultry Book (London: 1873). 3. Article on "Bantam," in Encyclopedia Americana, III (1941 ed.; New York: 1941), 215. 4. Dixon, E. S.: Ornamental and Domestic Poultry (London: 1848). 5. Pezard, A.; Sand, K., and Caridroit, F.: Archs. de biol. 36:541, 1926. 6. Greenwood, A. W., and Blyth, J. S. S.: JL Genet. 36:53, 1938. 7. Domm, L. V.: in Sex and Internal Secretions, ed. by E. Allen (2d ed.; Baltimore: 1939), 227. 8. Goodale, H. D.: Amer. Naturlst. 47: 159, 1913. 9. Morgan, T. H.: Proc. Soc. Exper. Biol. & Med. 13:31, 1915. 10. Morgan, T. H.: Proc. Soc. Exper. Biol. & Med. 15:3, 1917. 1 1. Morgan, T. H.: Carnegie Instn. Washington, Publn. no. 285, 1919. 12. Nonidez, J. F.: Jl. Anat. 31:109, 1922. 13. Nonidez, J. F.: Jl. Anat. 34:359, 1924. 14. Roxas, H. A.: Jl. Exper. Zool. 46:63, 1927. 15. Danforth, C. H., and Foster, F.: Proc. Soc. Exper. Biol. &: Med. 25:75, 1927. C. H. Danforth 165 16. Champy, C, and Demay, M. L.: Cpts. rd. Soc. de biol. 109:855, 1932. 17. Callow, R. K., and Parkes, A. S.: Jl. Exper. Biol. 13:7, 1936. 18. Gallagher, T. F.; Domm, L. V., and Koch, F. C: Ji. Biol. Chem. loo.xlvii, 1933. 19. Danforth, C. H.: Proc. Soc. Exper. Biol. & Med. 32:1474, 1935. 20. Torrey, H. B., and Horning, B.: Proc. Soc. Exper. Biol. & Med. 19:275, 1922. 21. Torrey, H. B., and Horning, B.: Anat. Recrd. 24:395, 1923. 22. Crew, F. A. E.: Proc. Roy. Soc. Edinburgh 45:252, 1925. 23. Danforth, C. H.: Jl. Exper. Zool. 65:183, 1933. 24. Parkes, A. S., and Selye, H.: Jl. Genet. 34:297, 1937. 25. Emmens, C. W., and Parkes, A. S.: Jl. Genet. 39:485, 1940. 26. Chu, J. P.: Jl. Genet. 39:493, 1940. 27. Danforth, C. H.: in Sex and Internal Secretions, ed. by E. Allen (Baltimore: 1932), 12. 28. Hill, R. T., and Parkes, A. S.: Proc. Roy. Soc. London, s.B 1 16:221, 1934. 29. Witschi, E.: Proc. Soc. Exper. Biol. & Med. 35:484, 1936. 30. Champy, C: Cpts. rd. Soc. de biol. 125:329, 1937. 31. Lyons, W. R., and Sako, Y.: Proc. Soc. Exper. Biol. & Med. 44:398, 1940. 32. Greenwood, A. W., and Blyth, J. S. S.: Proc. Roy. Soc. London, s.B. 1 19:97, 1935. 33. Willier, B. H., and Rawles, M. E.: Proc. Natn. Acad. Scis. 24:446, 1938. 34. Danforth, C. H.: Jl. Hered. 30:173, 1939. DESCRIPTION OF PLATE Fig. 1. Four successive feathers; a, b, and c definitely, d almost certainly, from the same follicle. These are from a golden Sebright ^vhich hatched Apr. 20 and was castrated May 27. The first two are cocky, the second two henny in character. la. Immature feather plucked Sept. 3. 16. Small section of the pinfeather which immediately replaced the above. This was plucked Sept. 17. The section is taken from well out toward the tip where barbules are lacking. The regions marked with crosses contain cells which degenerate without forming barbules. The six barbs shown are broAvn, not black as the photograph might suggest. ic. Section of the next pinfeather, which was plucked Oct. 2. The region is comparable to that shown in b. Beginning Sept. 19 and continuing till Oct. 5, about 0.5 mg. of testosterone propionate in oil was placed on the skin each day in the region of the follicle or on the developing pinfeather. It will be seen that under these conditions cells comparable to those marked with crosses in b de- veloped barbules and were invaded by pigmentophores, black to the left, broAvn to the right. id. Immature feather plucked Nov. 11 after the treatment had been resumed Oct. 10 and repeated daily from Oct. 12 to 17. Fig. 2. Feathers from a golden Sebright which hatched Apr. 20 and was cas- trated May 23. The upper row shows the effect of testosterone, the lower row the comparable effect of theelin. A total of 5.6 rag. of testosterone propionate in oil ^\as injected into the breast muscles during the period from Sept. 12 to 17. The upper ro^v shows the tips of three feathers which ^vere in different stages of development when the injections became effective. The one in the middle was affected in its proximal part, the one to right throughout its whole extent. (Only H — birds respond in this \vay; two hli birds of approximately the same size, age, and history treated in the same manner sho^ved no detectable effects on their plumage. All similarly treated specimens showed marked growth of comb and change in attitude.) In the second row are shown tips of three more feathers from the same bird plucked on Dec. 1 1 after 0.7 mg. of theelin in oil had been administered during a period of one ^veek when these feathers ^vere in different stages of development. Fig. 3. Feathers from a silver Sebright capon approximately one year old. On Aug. 2, about 1.4 mg. of testosterone propionate Avas embedded in, or slightly under, the skin near the center of a recently plucked area. The feathers shown are from replacements plucked Sept. 9. They are arranged as nearly as possible in the relative positions in which they grew and the api^roximate site of the testosterone implant is indicated by a cross. The four central feathers show varying degiees of "feminization"; the three peripheral ones and all the others that grew in this region failed to show any effect from the treatment. [1663 Plate 1 THE INFLUENCE OF HORMONES ON THE SEXUAL BEHAVIOR OF DOMESTIC FOWL By DAVID E. DAVIS Felloiu, International Health Division, Rockefeller Foundation Rio cle Janeiro, Brazil AND L. V. DOMM From the WHITMAN LABORATORY OF EXPERIMENTAL ZOOLOGY THE UNIVERSITY OF CHICAGO CHICAGO, ILLINOIS THE INFLUENCE OF HORMONES ON THE SEXUAL BEHAVIOR OF DOMESTIC FOWL* Introduction IN AN AiTEMPT to analyze the results obtained from a study on the sexual be- ha\ior of intersexual males produced by the injection of estrogens into incubating eggs (Domm and Davis'), a series of experiments was begun to de- termine the effect of androgens and estrogen^ on the behavior of the do- mestic fowl (Davis and Domm"). In the above investigation it was found that males in which one or both gonads were ovotestes showed various degrees of masculine behavior which roughly coincided with the degree of masculinity of the plumage. The plumage in turn is known to be an excellent indicator of the relative proportions of testicular and ovarian components comprising the gonad (Domnr). In brief, this study showed that males in which the plumage was essentially normal behaved in a definitely masculine manner while those in which the plumage was female, or nearly so, behaved in an essentially neutral manner. Gradations in behavior between these extremes were encountered. The observations to be discussed in this paper represent the beginning of an analysis of the function of certain hormones in the behavior of the fowl. Ex- periments designed to analyze the behavior of animals must necessarily be extensive because of the difficulty in isolating the numerous factors controlling behavior. For this reason it was considered desirable to conduct an initial series of general experiments, leaving the details to a subsequent, more extensive study. Therefore, the experiments reported at this time must be considered as an analysis of the general responses of capons and bilaterally ovariectomized poulards to androgens and estrogens; further refinements, such as the minimal amount of hormone necessary to initiate a particular response, the effects of learning or conditioning, etc., are reserved for a future time. Some incidental experiments upon males and sinistrally ovariectomized poidards are also con- sidered. Materials and Methods The brown Leghorn, a highly dimorphic breed of domestic fowl, was used throughout the experiments. The birds were kept in large pens at the Whit- man Laboratory poultry house and fed a standard ration of wet mash, dry mash, corn and greens, etc. (for details see Davis and Domm*). The experiments consisted of exposing the injected capons, roosters, and * This investigation was aided by a grant from the Dr. Wallace C. and Clara A. Abbott Memorial Fund of The University of Chicago. Grateful acknowledgment is made to Dr. Erwin Sch\\enk of the Schering Corporation for the testosterone propionate and a-estradiol (Progynon-B) and to Dr. J. A. Morrell of Squibb and Sons lor the stilbestrol used in these experiments. 1^2 Hormones and the Behavior of Fowl poulards to one or more of three stimulus situations. Depending upon the type of injection, the bird was exposed to a normal rooster, to a normal female, or to a dummy. In all cases possible the bird was tested in its own pen so that it was familiar with its surroundings. The normal roosters used were healthy individuals of the same age as the capons and in the prime of sexual vigor, except as mentioned in one experiment. The females also were normal vigor- ous birds which were good layers and receptive. The dummy was a normal hen stuffed and mounted in the pose of invitation to copulation (squatting). The routine experiment consisted of placing the test bird (rooster, hen, or dummy) in the empty pen and then introducing the experimental bird. Although the observer stood in the pen at a distance of a few feet, it is certain that his presence had no effect on the behavior. The reader is referred to the paper by Davis and Domm* for a detailed description of the normal sexual behavior of the fowl. The birds were injected daily at about 11:00 a.m. In a few days they be- came tame, and readily submitted to the treatment so that there is little if any possibility that the ordeal of injection influenced their behavior. The reader is referred to the table which gives a summary of pertinent data. Observations The series of injections was designed to indicate the action of androgens and estrogens in controlling behavior. Since in general the end result is the im- portant point, description of the developmental stages will be omitted for the most part in this paper. 1. The Effect of Androgen in the Capon. Two capons under the influence of testosterone propionate copulated repeatedly with both the dummy and females. Bird 149M was first observed to crow on the eleventh day and to tread on the fourteenth day of injection. Bird 153RM copulated first on the seventh day and crowed on the tenth. Since the birds were not under continuous ob- servation it is quite likely that they crowed several days before the dates men- tioned.* A third bird, 143R, which received the lowest dosage, was very excitable during the experiments and this excitability had not noticeably vanished six months after injections were stopped. He never copulated perfectly although, in contrast to a normal capon, he perceived the dummy. Our evidence is not sufficient at present to state that the dosage received in this case is below the threshold necessary for copulation. 2. The Effect of Estrogens m the Capon. All three capons copulated under the influence of a-estradiol (Progynon-B). The action was brief but complete; wing-fluttering and movements of tail and cloaca were typical. The birds getting the higher dosages (149M and 153RM) mounted, although weakly, on the sixteenth day (Oct. 7) of injection. Bird 170Y mounted two days later. How- ever, these birds, when compared with the testosterone-treated individuals, did * We have since observed crowing in 48 hours and treading within 90 hours in capons re- ceiving daily injections of testosterone. Davis and Domm i^o not show such intense Hbido and did not run as rapidly toward introduced fe- males. There was no crowing or "tidbitting." A very weak "waltz," consisting merely of lowering the wing, occurred occasionally. "Waltzing," which is used by the male to induce the newcomer to indicate its sex by either squatting (female) or raising the neck hackle (male), consists of a lowering of the wing nearest the newcomer and a kicking of the leg on the same side against the lowered wing. It must be remembered that "waltzing" does occur, although rarely, in capons which appear to be devoid of testicular tissue. "Tidbitting" is a characteristic coaxing, clucklike call accompanied by pecking at or tap- ping the ground or floor of the pen with the beak, which usually holds a morsel. This pattern is a substitute behavior, usually observed when a male is frustrated in his attempt to copulate. Since the effect of estrogen in causing the capons to copulate was heterodox, another series was begun in order to check the results. As indicated in the table, three birds were injected with a-estradiol and three with stilbestrol. Of the estradiol birds, i6gRY copulated on the sixth day, igsYY on the fifth day, and 193MY on the fourth day of injection. Of the stilbestrol birds, 382MM copulated on the fourth day, and igoYY on the fifth day. 191RR was a very nervous bird and never copulated. Sexual libido normally mani- fested itself in the estrogen-injected capons by their frequent attempts to tread one another, which occasionally succeeded, as for example when 382MM raped 191RR on the afternoon of November 5. When a female was released in the pen with all the capons, a veritable melee resulted from the attempts of the birds to copulate. There is a possibility that the capons became refractive to estrogen, for some of the birds which had copulated regularly did not copulate in the last few trials at the end of the experiment. When exposed to a normal rooster the estrogen-injected capons invariably fled. They never squatted for copulation and were unreceptive and struggled whenever the rooster forced copulation. The position of the dummy is important in the reaction of the estrogen- injected capon. If the dummy is facing him, he raises the hackle and begins to fight. If, on the other hand, it is not, he copulates. For example, on October 24, capon 170Y which had received 0.50 mg. a-estradiol daily for thirty-three days approached the dummy from the front, raised the hackle, and pecked the head. In walking away he happened to circle the dummy. He then ap- proached it from the rear and copulated. 3. The Effect of Estrogen in the Rooster. The two normal roosters which re- ceived stilbestrol gradually lost their aggressiveness and stopped crowing on the tenth day of injection. Nevertheless, copulations continued to the end of the experiment (47 days). The behavior pattern was complete but was per- formed in a lethargic, listless manner. The comb decreased practically to the castrate condition in each, and when autopsied on August 5 the testes of both were in a markedly regressed condition (macroscopically), weighing 75 and 110 gm. per bird. A third bird became sick soon after the start of the expcri- 174 Hormones and the Behavior of Fowl ment and consequently injections were discontinued. Stilbestrol acts the same as a-estradiol in producing feminization of growing feathers. These experiments therefore indicate that androgen in the male sex in- duces crowing and possibly "tidbitting," and increases the frequency of or actually induces "waltzing," and that both androgens and estrogens elicit the patterns of copulation. Summary of Data on Various Types of Injection Sex* His tory Hormone Amount, mg. per day Injec- tions, daysj Bird No. Hatching date Castration datet Results I43R c 5- 6-40 6- 6-4O T. P. 1.25 26 Imperfect copula- tion; very nervous I49M c 5- 6-40 6- 6-40 T. P. 2.50 26 Copulated, crowed 153RM c 5- 6-40 6- 6-40 T. P. 3-75 26 Same 170Y c 5- 6-40 6- 6-40 a-Estra. ■50 43 Copulated I49M c 5- 6-40 6- 4-40 a-Estra. 1 .00 43 Same I 53RM c 5- 6-40 6- 6-40 a-Estra. 1.50 14 Same (imperfectly; sick) I92YY c 5- 6-40 6- 4-40 a-Estra. 1 .00 14 Copulated 1 93 MY c 5- 6-40 6- 4-40 a-Estra. I. i;o 14 Same 169RY c 5- 6-40 6- 4-40 a-Estra. 2.00 II Same 382RR c 5- 6-40 6- 6-40 Stilb. 1 .00 14 No copulation; very nervous 382MM c 5- 6-40 6- 6-40 Stilb. 1.50 14 Copulated 382YY c 5- 6-40 6- 4-40 Stilb. 2.00 14 Same 169RY c 5- 6-40 6- 7-40 Control Behavior neutral 170Y c 5- 6-40 6- 6-40 Control Same 180R c 5- 6-40 6- 7-40 Control Same 378 R 5- 6-40 Stilb. 2. 50 47 Copulated if 9 squat- ted; stopped crow- ing and "waltzing" 323PU R 5- 6-40 Stilb. 5.00 47 Same 320BW R 5- 6-40 Stilb. 7.50 25 Sick (died of roup) 118PU BP 6-20-38 8-12-38 T. P. 1.25 60 Crowed,"waltzed,"no copulation 132WUR BP 6-20-38 8-12-38 T. P. 2.50 60 Same 128RU BP 6-20-38 8-16-38 T. P. 3-75 60 Same ii8PU§ BP 6-20-38 8-12-38 a-Estra. .50 50 Squatted, received copulation 132WUR BP 6-20-38 8-12-38 a-Estra. 1 .00 43 Same 128RU BP 6-20-38 8-16-38 a-Estra. 1.50 5° Same (see text) Davis and Domm 175 Summary of Data on Various Types of I>fjECTio>f {Continued) History Sex* Hormone Amount, mg. per day Injec- tions, dayst Bird No. Hatching Date Castration Datet Results I21G BP 6-20-38 8-16-38 Control Behavior neutral I35PB SP 6-20^38 8- 2-38 T. P. 1.25 26 Increased frequency of crowing and "waltz- ing," no copulation I20M SP 6-20-38 7-3^38 T. P. 2.50 26 Same II3R SP 6-20-38 7-29-38 T. P. 3-75 26 Same iiiU SP 6-20-38 7-28-38 Control Crowed, "tidbitted," "waltzed" weakly; no copulation I26BW SP 6-20-38 8- 1-38 Control Same I27Y SP 6-20-38 8- 1-38 Control Same * C, capon; R, rooster: BP, bilaterally ovariectomized poulard; SP, sinistrally ovariectomized poulard. t Date of final operation. Except in sinistral poulards, castrations were performed in two operations. X Experiments performed from June to November 1941. § The bilaterally ovariectomized poulards used in this experiment are the same ones as were previously used in the above experiment on the effect of androgen in bilaterals. An interval of approximately one month elapsed between the two experiments. 4. The Effect of Androgen in Bilaterally Ovariectomized Poulards. A bilater- ally ovariectomized poulard is one in which the left ovary has been completely removed following or preceding the complete removal or destruction of the right rtidimentary gonad. Such individuals are, therefore, gonadless and re- semble the capon in many important respects. Like the capon they are neutral in behavior (Domm^'^). Under the influence of androgen bilaterally ovariectomized poulards be- came very aggressive and fought violently. The birds were first ob,served at- tempting to crow on the sixteenth day of injection. They began to crow in a perfect manner a few days later and continued to do so throughout the experi- ments. Within three weeks all three poulards "waltzed" at introduced females or the dummy. The "waltzing," although not as vigorous as that of a male, was nevertheless identical in behavior pattern. These three poulards, however, did not copulate with receptive females or the dummy at any time, even though the experiments were continued for sixty days. The poulards gave all females as well as the dummy vicious pecks about the head, even after a "waltz." Females which squatted perfectly, thereby inviting copulation, were often injured by the violence of these pecks. 5. The Effect of Androgen in Sinistrally Ovariectomized Poulards. Sinistrally ovariectomized poulards are individuals in which the left ovary only has been removed. Following this operation the rudimentary right gonad undergoes hypertrophy, usually developing into a tcstis-like organ whose secretions are mixed, having both androgenic and estrogenic effects (Lillie'). The androgenic 176 Hormones and the Behavior o£ Fowl effects almost invariably appear soon after the operation and may bring about a more or less complete masculinization of sexual characters. The estro- genic effects usually become evident some time later and are chiefly mani- fested by the plumage (Domm^). Following daily injections of testosterone, sinistrally ovariectomized pou- lards became noticeably more aggressive. They were observed to crow more frequently than uninjected controls and likewise displayed more interest in the female, notably evidenced by the greater frequency of "waltzing." How- ever, like their uninjected control mates, they were never observed to copulate, but merely "waltzed" and stopped or "waltzed" and pecked. These individuals definitely became more masculine in their general reactions under the influ- ence of testosterone but apparently they were unable to copulate, at least under the stimulus conditions prevailing in these experiments. 6. The Effect oj Estrogen in Bilaterally Ovariectomized Poulards. These in- dividuals, normally lacking aggressiveness, seemed to become more timid and even lethargic following daily injections of estrogen; they completely ignored the dummy and normal females in all tests. Bird 132WUR, eighteen days after the beginning of injections, squatted for a normal rooster and received copula- tion. These experiments at this date were handicapped by the lack of a vigor- ous rooster, since all the males were in molt. It is possible that the injected poulards would have squatted regularly at this time for an aggressive, vigorous rooster in sexual prime. Hence, a rooster was brought into full sexual ag- giessiveness by daily injection of 2.0 mg. of testosterone. In the presence of this rooster, bird 132WUR squatted regularly and 118PU squatted on the forty- second day of injection. Bird 128RU did not squat until November 10, after it had been severely beaten by its cage mate. Although she received copulation subsequently when exposed to the rooster it is not known whether she was receptive or merely too weak to resist. On several occasions an estrogen injected bilaterally ovariectomized pou- lard squatted for an estrogen-injected capon and received copulation. These birds were all identical in plumage, lacked any vestige of gonads, and had received the same injections. It seems apparent from these observations that androgen induced crowing and "waltzing" but did not induce copulation in the bilaterally ovariecto- mized poulard. These individuals became definitely more active and aggres- sive in the presence of the introduced female or dummy as a result of the injections. It is also evident that androgen increased the frequency of crowing and "waltzing" in sinistrally ovariectomized poulards but, as was the case in the bilaterals, it did not induce them to copulate, at least under the various stimulus situations provided. Estrogen, on the other hand, apparently induces only the squatting pattern in the bilaterally ovariectomized poulard. It will be noted that this is in strik- ing contrast to the effect of estrogen in the capon. One point deserves particu- lar mention, bilaterals receiving estrogen completely ignored the dummy or Davis and Domm ^^^ other females. In contrast, those receiving androgen definitely noticed the dummy and reacted toward it in much the same way they reacted toward normal hens. Discussion The action of androgens and estrogens in the control of behavior has been studied in few species. In the domestic fowl it has been repeatedly shown that testosterone will produce crowing in both males (Benoit') and females (Allee, CoUias and Lutherman"; Hamilton and Golden'"). It is important to note that these females did not copulate. In young male chicks Hamilton"^ produced crowing at ten days of age although the females did not crow in fifty-five days. Later Hamilton and Dorfman"^ caused male chicks to crow sixty hours after hatching. In addition, Allee, Collias and Lutherman" observed crowing in four-month-old pullets. These authors also noted a somewhat abbreviated "waltz" or courtship pattern in two of their females. By stimulating develop- ment of gonads precociously with pituitary hebin, Domm and Van Dyke"" and Domm'^ observed crowing in young cockerels at nine days of age and treading at thirteen days. These authors observed little if any effect on the behavior of similarly treated young pullets even though there was sufficient androgen produced, by the hypertrophied medullary component of the left ovary, to bring about masculinization of head furnishings comparable to that of treated cockerels. In female canaries Leonard" induced singing by injection of testosterone and Shoemaker" thus produced singing and covirtship behavior as wtII as an increased aggressiveness, but no copulations. In nondomesticated birds Noble and Wurm'* found that in the black-crowned night heron testost- erone will induce masculine sexual behavior in adult females and even in month-old chicks. The patterns consisted of territory defense, nest building, courtship ceremonies, and copulation on one occasion. In the laughing gull Noble and Wurm" found that testosterone produces those behaviors common to both sexes as well as those of the male. In mammals under the influence of testosterone BalP found that female rats performed masculine copulatory patterns and Beach"' observed that male rats performed female mating behavior. The latter author concluded that the stimulus situation is the selective factor that determines which type of response will be manifested under the conditions of great excitability produced by testosterone. Similarly in the lizard Aiiolis carolinensis, Noble and Greenberg" found that testoster- one elicits both male and female behavior. Evans,^"' studying Anolis carolin- ejisis, found that female lizards which had been castrated fought much more than normal individuals and that the injection of ovarian hormone inhibited the fighting. The effect of estrogen on behavior is less striking. In chickens Allee and Collias^ found that one poulard, sinistrally ovariectomized by L. V. Domm, squatted under the influence of estrogen. Noble and Wurm'* found that no behavior of the female black-crowned night heron is due to estrogen. These authors also found" that the injection of estrogen into either sex of the laugh- 178 Hormones and the Behavior of Fowl ing gull will induce "food-begging," a behavior pattern which is necessary for coition. In mammals BalP" found that estrogen could produce the feminine behavior patterns in castrated male rats, but that these individuals continued to perform the masculine copulatory patterns. That there is some difference in the potentialities of the sexes is shown by the fact that more estrogen is required to produce lordosis in males than in females and that females did not show masculine tendencies under the influence of estrogen. The evidence so far accumulated indicates that both sexes of the domestic fowl possess the potentialities for performing certain masculine behavior pat- terns. Crowing, "waltzing," and "tidbitting" occur in males and females under the influence of testosterone. Crowing certainly is induced by androgen, while "waltzing" and "tidbitting," if not actually induced by testosterone, are at least increased in frequency. Further research upon the stimulus situation which will induce a capon to "waltz" is necessary before a solution of this problem can be reached. Fighting and aggressiveness, although occurring in capons, is very greatly increased by the injection of testosterone. That there may be a basic quantitative difference between the sexes is indicated by the observation that capons crowed sooner than poulards under the influence of androgen. In contrast to the ability of both sexes to perform certain behavior patterns is the observation that masculine copulatory behavior was not induced by the injection of testosterone into poulards. This is surprising because in some species of birds (Shoemaker;" Noble and Wurm^^) the females are reported to mount, and some poultrymen say that hens sometimes tread one another. Also, there is one case on record (Domm^) in which a sinistrally ovariectomized poulard is known to have performed the male copulatory act. Whether other stimulus situations than those employed in our experiments would elicit copu- lation in a testosterone-injected poulard is not known. In any case, we can confidently state that there is a sexual differential, since under the experi- mental procedures employed the capons copulated whereas the poulards did not. It is of further interest that both androgen and estrogen induced the copulatory behavior pattern in our capons. Apparently in this case the hor- mones are not specific in their action on the nervous system. These results would seem to suggest that the female chicken may not have the proper nerv- ous patterns for the masculine copulatory behavior save in very exceptional cases. It is worthy of note that the behavior relative to reproduction may be divided into two types. The primary patterns are related to the actual insemi- nation. The secondary patterns (epigamic, Huxley^) include such actions as singing (crowing) and courtship ("waltzing"). It is perhaps more than coinci- dence that the typical male primary patterns appear to be sex limited but can be activated by either androgen or estrogen, while, on the other hand, the secondary patterns are induced in both sexes by androgen. Further research is necessary to determine the validity and application of this generalization. The results of this series of injections throw some light on the problem of Davis and Domm i ^n the inhibition of behavior patterns. Many female birds show mascuHne be- havior in the fall, such as territory holding (Michener and Michener;" Lack^) or singing (Bullough and Carrick""), but cease these activities in the spring. The obvious interpretation of this sequence of events is that estrogen, when it appears in the spring, inhibits the expression of masculine behavior in the same manner as it inhibits the expression of the masculine plumage in certain species. The results of our experiments suggest the likelihood that these masculine behavior patterns are produced in the fall by androgen and not by estrogen. However, experiments consisting of the simultaneous injection of androgen and estrogen are required to test the hypothesis of inhibition. Summary 1. In order to analyze the relation of certain endocrines to sexual behavior patterns in the domestic fowl, a series of injections was performed on capons, sinistrally and bilaterally ovariectomized poulards, and normal roosters, using testosterone propionate, a-estradiol (Progynon-B), and stilbestrol. The behavior was tested by exposing the injected bird to a normal rooster, to a normal hen, or to a stuffed female dummy mounted in the pose of invitation to copulation (squatting). 2. Two out of three capons receiving testosterone copulated, croAs'ed, "waltzed," and "tidbitted." The six capons receiving a-estradiol and two of the three receiving stilbestrol copulated in the masculine manner. None of them were known to crow or "tidbit." Two normal roosters receiving stil- bestrol stopped crowing but continued to copulate. 3. Three bilaterally ovariectomized poulards receiving testosterone crowed, "waltzed," and became aggressive, but never copulated throughout the sixty days of injections. Three sinistrally ovariectomized poulards reacted similarly. Two out of three bilaterally ovariectomized poulards receiving a-estradiol squatted for the rooster. In the stimulus situations employed in these experi- ments copulation was not induced in poulards by injections of testosterone. 4. Only in the male can the behavior patterns of copulation be induced by both androgen and estrogen, while the patterns of crowing, "waltzing," and "tidbitting" are under the control of androgen in both sexes. Addendum Since this manuscript was submitted an important report has appeared (Zitrin'") describing induction of male copulatory behavior in a hen follow- ing administration of male hormone. This bird, one of two treated white Leghorn pullets, received a pellet of testosterone propionate, weighing ap- proximately 41 mg., when it was five months old. The bird was observed to tread a hen one hundred thirty-eight days later. Thereafter it was tested fre- quently with a squatting hen and observed to copulate in the typical masculine manner. "Waltzing" was seen only occasionally and was never very vigorous, while crowing was heard but once after the bird had been treaded by a male. 1 8o Hormones and the Behavior o£ Fowl Male copulatory behavior was exhibited for approximately one and a half months after which the bird began to lay fertile eggs. In order to interpret certain phases of the experiments reported in the pres- ent paper, a new series (Domm, Davis, and Blivaiss^) was undertaken which closely parallels the experiment of Zitrin in certain respects. Nine brown Leghorn pullets, divided into three groups of three pullets each, received daily injections of testosterone propionate. The first group (112 days old) received 0.50 mg., the second, which was the same age, received 1.00 mg. and the third (138 days old) received 1.50 mg. daily. All birds were tested and observed daily during the first two months and usually on alternate days thereafter. The group receiving 0.50 mg. was injected for one hundred twenty-one days. One of these crowed, circled, and "waltzed"; another crowed biu never showed any particular interest in either sex; while the third was not known to crow but began to chase, circle, and "waltz" shortly before injections were discontinued. The other two groups of six pullets are still under observation and have now received daily injections of the amounts indicated for more than one hundred eighty days. All of these are known to crow, chase, circle, and "waltz" and sev- eral have shown weak mounting attempts. Only one of those receiving the highest dosage has actually been observed to mount. This bird grabbed and mounted a squatting female on the one hundred forty-sixth day in the typical male manner. However, copulation did not take place and a repetition has not occurred since, though several weak attempts at mounting have been observed. In all groups, crowing, chasing, circling, and "waltzing" became quite common once these reactions appeared. In attempting to evaluate these results one is likely to raise a question con- cerning the relative efficacy of the two procedures employed in administering the hormone, since it is known that a constant source of hormones is necessary for the development and maintenance of many sexual characters. However, until more data are available concerning the development and maintenance of the various sexual-behavior patterns in the fowl the sporadic occurrence of treading females, whether normal or the result of various experimental pro- cedures, must be regarded as exceptional. REFERENCES 1. Domm, L. V., and Davis, D. E.: Proc. Soc. Exper. Biol. & Med. 48:665, 1941; Anat. Recrd. 8i(supp.):6i, 1941. 2. Davis, D. E., and Domm, L. V.: Proc. Soc. Exper. Biol. & Med. 48:667, 1941; Anat. Recrd. 8i(supp.):io4, 1941. 3. Domm, L. V.: in Sex and Internal Secretions, ed. by E. Allen (2d ed.; Baltimore: 1939), 227. 4. Davis, D. E., and Domm, L. V.: In press. 5. Domm, L. V.: Jl. Exper. Z06I. 48:31, 1927. 6. Domm, L. V.: Biol. Bull. 56:459, 1929. 7. Lillie, F. R.: Jl. Exper. Z06I. 48:175, 1927. 8. Benoit, J.: Archs. de zool. exper. et genrl. 69:217, 1929. 9. Allee, W. C; Collias, N., and Lutherman, C. Z.: Physiol. Zool. 12:412, 1939. 10. Hamilton, J. B., and Golden, W. R. C: Endocrinology 25:737, 1939. Davis and Domm 1 8 1 11. Hamilton, J. B.: Endocrinology 23:53, 1938. 12. Hamilton, J. B., and Dorfman, R. I.: Endocrinology 24:71 1, 1939. 13. Domm, L. V., and Van Dyke, H. B.: Proc. Soc. Exper. Biol. & Med. 30:319, 1932. 14. Domm, L. V., and Van Dyke, H. B.: Science (n.s.) 77:456, 1933. 15. Domm, L. V.: Cold Spring Harbor Sympsa. Quantit. Biol. 5:241 , 1937. 16. Leonard, S. L.: Proc. Soc. Exper. Biol. & Med. 41:229, 1939. 17. Shoemaker, H. H.: Proc. Soc. Exper. Biol. &: Med. 41:299, 1939. 18. Noble, G. K., and Wurm, W.: Endocrinology 26:837, 1940. 19. Noble, G. K., and Wurm, ^V.: Anat. Rccrd. 78(supp.):50, 1940. 20. Ball, J.: Jl. Compar. Psychol. 29:151, 1940. 21. Beach, F. A.: Endocrinology 29:409, 1941. 22. Noble, G. K., and Greenberg, B.: Proc. Soc. Exper. Biol. &: Med. 47:32, 1941. 23. Evans, L. T.: Pedagog. Seminary & Jl. Genet. Psychol. 48:217, 1936. 24. Evans, L. T.: Physiol. Zool. 10:456, 1937. 25. Allee, W. C., and Collias, N.: Endocrinology 27:87, 1940. 26. Huxley, J. S.: Proc. Vllth Internatn. Ornithol. Congr., 1930:107, 1931. 27. Michener, H., and Michener, J. R.: Condor 37:97, 1935. 28. Lack, D.: Proc. Zool. Soc. London, s..\. 109:169, 1939. 29. Bullough, W. S., and Carrick, R.: Nature (London) 145:629, 1940. 30. Zitrin, A.: Endocrinology 31:690, 1942. 31. Domm, L. V.; Davis, D. E., and Blivaiss, B.: Anat. Recrd. 84fsupp.):3i, 1942. PITUITARY GONADOTROPHINS By HEINZ FRAENKEL-CONRAT, CHOH HAO LI AND MIRIAM E. SIMPSON From the INSTITUTE OF EXPERIMENTAL lUOI.OGY UNIVERSITY OF CALIFORNIA, BERKELEY, CALIFORNIA < PITUITARY GONADOTROPHINS WITHIN THE past decade various gonadotrophic hormones have been puri- fied considerably. E\idence for purity of the active principles from human pregnancy urine and serum of pregnant mares has been reported.'"* Purification of a complete gonadotrophin from pituitary extracts has not been reported, while in several laboratories fractionation of such extracts has yielded two separate gonadotrophic principles of a high degree of activity.""'" The biological effects of these two principles are similar though not identical in character with those ascribed ten years ago by Fevold and his co-workers" to their follicle-stimulating (FSH) and luteinizing (LH) hormone respectively. The close resemblance in the activity of the purified preparations obtained by different methods in a number of laboratories does not lend support to the assumption that these two active principles may represent artifacts or split products of a macromolecule. It is recognized, however, that the isolation of these proteins from the pituitary does not prove that they actually are secreted in this form. Evidence has been presented that the hormones liberated by intact glandular tissue are quantitatively and qualitatively superior to those extracted from stich tissue.^* However, the concentration of pituitary gonado- trophins in the blood stream is so low that no atcempts to purify and, possibly, fractionate these hormones, have as yet been reported. The fact that the type of gonadotrophic activity in the body fluids of normal (nonpregnant) animals may show great variations within the life cycle of the animal, primarily con- trolled by its gonads, tends to favor the assumption of two hormones actually being secreted in varying relative amounts by the pituitary. With the study of these circulating hormones being delegated to the future, a review of the properties of the gonadotrophins isolated from pituitary extracts will be attempted here. Physicochemical Properties of FSH and ICSH (LH) (Foil iciest i mulatitig hormone mid interstitial-cell-stiinidaliug honnone) All gonadotrophins are glycoproteins; they can be effectively separated from inert protein and most other hormones by their stability and solubility in alcohol-water or acetone-water mixtures. Extraction of acetone-dried sheep glands with 40 per cent alcohol has proven most successful.''" Methods of fractionation of the gonadotrophic complex are based on the fact that the FSH is more soluble in salt solutions than the ICSH. Thus, repeated precipita- tion of the proteins with ammonium sulfate at half and two-thirds saturation, as well as fractionation with sodium sulfate at pH 4.4, achieves separation of the two principles. From the lower salt fractions a pure protein was isolated which contained the ICSH activity.'""'-""''' Comparison of the final products obtained from sheep'""" and pig piiuitaries"" has shown these to be quanti- i86 Pituitary Gonadotrophins tatively and qualitatively similar in biological activity but quite different chemically (see table i). The follicle-stimulating factor has not yet been obtained in pure form. Active fractions are soluble in water and half saturated ammonium sulfate; they contain almost twice as much carbohydrate as the ICSH."'^" Comparison of FSH from different species has to be postponed until final purification will have been achieved. At the present it can only be stated that the best sheep FSH is considerably more active and appears to be more stable than is pig TABLE 1 Physicochemical Characteristics of ICSH from Sheep and Hog Pituitary Glands* Determinations Molecular weight Sedimentation content, S X lo'^. Isoelectric point, pH Tyrosine, per cent Tryptophane, per cent Mannose, per cent Hexosamine, per cent Sheep 40 , 000 3-6 4. 60 4.50 1 .00 4-45 5.86 Hog 90 , 000 5-4 7 ca. 3 2 45 80 * See ref. 2, 10, 11, 12, 14, 15. FSH.*'^'^ The latter has been claimed to be "biologically pure,"^ that is, free from all active contaminants, while the best sheep FSH fractions may contain some ICSH, but no other known hormones. A considerable amount of work has been done to study the effect of specific reagents on the hormonal activity of the gonadotrophins. Thus, some proteo- lytic enzymes were found to inactivate preferentially either ICSH (LH) or FSH. Crude trypsin was shown to destroy nearly all the ICSH but not the FSH,^'-^ while pepsin leaves intact only the ICSH;^^ on the other hand, ptyalin (saliva) destroys FSH but not ICSH.^ Apparently the high carbohydrate con- tent of the FSH is in some way, possibly only structurally, essential for its activity. Rapid inactivation by treatment with ketene^^ and nitrous acid may be interpreted as evidence for the essentiality of free amino groups for the activity of the gonadotrophins. Cysteine also inactivates these hormones, although not as readily as other proteins,^' ^ thus favoring the assumption that the integiity of some disulfide bonds which are not easily reduced is essential for hormonal activity. Biological Properties of ICSH As the name indicates, this hormone stimulates the interstitial tissue of the gonads of either sex. Its ability to effect repair of this tissue, which atrophies characteristically after hypophysectomy, has been used as a test method in female rats. By this method 5-10 ix%. of the hormone can be detected. Other Fraenkel-Conrat, Li, and Simpson 187 effects in the female become evident only in combination with P'SH and will be discussed later. In male rats, stimulation of the interstitial tissue leads to androgen secretion, thus offering a functional test method for this hormone. It has been shown, however, that the ability of all pituitary extracts and frac- tions to stimulate androgen secretion is very poor compared with that of circulating hormones.®'^ Thus, most authors (with the exception of Fevold and co-W'Orkers) find the seminal vesicles of immature normal or hypophy- sectomized rats very insensitive to stimulation by pituitary preparations.* The observation that the ventral prostate of immature hypophysectomized rats (2 days p.o.) was much more sensitive to the secreted androgens than other parts of the accessories has supplied a reliable functional-test method for the interstitial-cell-stimulating hormone."" Other effects of this hormone may be less specific. Thus, it increases the testis weight of normal and hypophysectomized rats and, to a greater extent, those of immature pigeons and chicks, an effect which is also produced by FSH. Recently ICSH was also found to favor placentoma formation and pro- long pregnancy in normal rats,^° phenomena which are probably consequent to stimulation of the pituitary by the ICSH. Another type of biological activity can be produced with ICSH, that is, that of antagonism. Upon intraperitoneal administration, this hormone will de- crease the activity of other gonadotrophins (such as pregnant-mare serum) simultaneously administered.^^ Biological Properties of FSH As the name indicates, the primary effect of this hormone in the female is the stimulation of the development of follicles. In hypophysectomized rats, only 2-3 jxg. of the most active preparations from sheep pituitaries are needed for this effect." Pig FSH fractions which have been described as "biologically pure," that is, free from other active principles, are considerably less active.^ The follicular giowth elicited with purified FSH is associated with only lim- ited increases in the weights of the ovaries, which rarely exceed 60 mg. Follicles produced with FSH alone seem to secrete less estrogenic substances than those produced by unfractionated extracts, as judged by uterine de- velopment. According to Creep and his associates, "biologically pure" pig FSH was found to result in no uterine stimulation, even at very high dosage.^ Sheep FSH has not been obtained similarly free from estrogenic properties." At sufficiently high levels (ten- to fiftyfold the MED) in hypophysectomized rats it produces estrus, corpora lutea, and repair of the interstitial tissue in female and ventral prostate growth in male rats. From these findings it could be con- cluded that the best sheep FSH fractions still retain 10-20 per cent ICSH. The action of FSH preparations on interstitial tissue was found to differ, however, * ICSH maintains the accessories of adult male rats after hypophysectomy but docs not produce, in any type of rat, hypertrophy comparable to that elicited with secreted gonado- trophins. i88 Pituitary Gonadotrophins from that of ICSH in being more pronounced upon subcutaneous than intra- peritoneal administration. Furthermore, it has been shown that purified FSH upon intraperitoneal administration antagonizes other gonadotrophins (even subcutaneously administered FSH). Such antagonistic phenomena can be produced with ecpial or even smaller amounts of FSH than are needed of ICSH. It thus appears unlikely that such antagonism is due to contamination with ICSH; antagonism probably is an inherent property of each gonado- trophin.^^ These findings suggest the possibility that a limited degree of ICSH-like properties may be characteristic of sheep FSH. This could be proven only by isolation of physicochemically homogeneous FSH still showing these diverse activities. Such relationships between hormones have been shown to exist in the case of the adrenal steroids, several of which may combine high activity in one respect with low activity in another direction, in which another hor- mone of a slightly different structure may be highly active. Furthermore, the gonadotrophin from pregnant mares appears to be a single protein while showing both FSH and ICSH properties.^ In hypophysectomized rats FSH causes increases in testis weights, which are more pronounced than those produced by similar amounts of ICSH. His- tological examination indicates stimulation of the tubular development, although it appears that spermatogenesis can be furthered much more effec- tively if FSH is combined with ICSH. Testosterone alone has been shown to be effective in both maintaining and repairing testicular function of hypophy- sectomized rats; the gametrotropic effect of FSH is difficult to explain on the basis of ICSH contamination and concomitant testosterone production, in view of the atrophic condition of the accessories of rats treated with effective levels of FSH. Synergism It has been recognized for a long time that ICSH shows pronounced activity in augmenting and modifying the action of FSH when both are given sub- cutaneously. A reinvestigation of these synergistic effects, using purified hor- mones, has led to the following conclusions: In female rats, immature normal or hypophysectomized, subcutaneously injected ICSH augments the effects of FSH on the ovaries and in particular the secondary effects on the uterus.^' ^^ Ovarian-weight augmentation rarely exceeds loo per cent, in contradistinction to that produced by combined administration of FSH with "secreted" inter- stitial-cell-stimulating substances, such as the pregnancy urine principle. On the other hand, uterine "augmentation" becomes quite pronounced when less than 5 units of sheep FSH (alone insufficient to effect estrogen secretion) is combined with about i unit of ICSH. This is all the more surprising since 5 units of ICSH would be needed to produce any effect on the interstitial tissue in the same rats upon subcutaneous administration. Actually, under favorable conditions, as little as one-fifth to one-tenth of a unit of ICSH may be detected by its synergistic effect with 3-5 units of FSH on the uterus of hypophysecto- Fraenkel-Conrat, Li, and Simpson 189 mized rats. Histological study of the ovaries of such doubly treated rats reveals no indication of the presence of ICSH other than by its augmentation of the follicle stimulating action of FSH. When ICSH is combined with higher doses of FSH, however, it may lead to luteinization of the follicles. This phenome- non led to its discovery and has prompted the name "luteinizing hormone (LH)."^ Since sheep FSH alone will cause luteinization of follicles at suffi- ciently high doses, it is difficult to decide whether the luteinizing action of ICSH is a specific one or is due only to its FSH augmenting action previously discussed.* In particular, after it was shown that similar augmentation and luteinization effects could be produced on combining FSH with inorganic salts or inert proteins, doubt was cast on the existence of this second gonado- trophin, "LH." While it appears well established that FSH can be augmented in its effect by nonspecific agents, the above described augmentation phe- nomena of FSH produced by ICSH are most certainly due to a specific syner- gism of the two substances, for the following reasons: (i) These effects can be produced by injection at different sites of the body of the two gonadotrophins, whereas nonspecific augmentation can be produced only by administration of the mixture, and are most probably due to delayed absorption of the hor- mone;^''" (2) less than i /xg. of ICSH augments the effect of a few microgiams of FSH, whereas very large doses of nonspecific agents are needed to produce stich augmenting effects; (3) heat inactivation of ICSH destroys also its capac- ity to augment FSH. It may be hoped that as a result of the purification of ICSH and with the recognition of its intrinsic effect on the interstitial tissue (which is not given by salts and inert proteins), arguments against its existence based on augmentation and luteinization phenomena will not be voiced any longer. It has been mentioned that the ovarian weights produced by FSH (in normal immature or hypophysectomized rats) rarely exceed 60 mg. after three days of treatment; and that added ICSH augments these weights to no more than double. By administering purified FSH and ICSH in the form of tannates, however, ovaries ranging from 200 to 300 mg. can be produced under the same conditions, similar to those produced with crude extracts. It thus appears possible that the inert proteins in crude extracts may exert a nonspecific aug- menting action which can be replaced by the use of tannic acid. Whereas augmentation of the effects of FSH by ICSH in females are well recognized, augmentation of ICSH by FSH does not appear to occur; on the contrary, marked follicular stimulation may mask the action of ICSH on the interstitial tissue. In the male rat no clear evidence for synergism between purified FSH and ICSH has been found.f Interesting parabiotic experiments are reported in the * Usually this luteinization of follicles obtained by combiniii;4 low or moderate levels of FSH with ICSH is thecal luteinization in contrast to the granulosa lutcini/ation characteristic of high levels of purified sheep FSH. t Nonspecific augmentation of the effect of ICSH on the \fnlial prostate of hypophysecto- mized rats has been observed, using copper sulfate. igo Pituitary Gonadotrophins literature in which a male and a female rat were subjected to the control of the same pituitary.^ In the female, follicular and uterine stimulation were evident; in the male, testis and accessory-organ growth occurred. Duplication of these conditions and effects by treating males and females wath a known FSH-ICSH mixture has shown that these cannot be interpreted as evidence for the secretion by the pituitary of a single gonadotrophin.^* The observed phenomena are better explained by the fact that, in the female, ICSH aug- ments the action of the FSH while, in the male, only its intrinsic hormono- trophic action becomes evident. Conclusions An attempt has been made to summarize and coordinate briefly the results of investigations carried out in this and other laboratories within the last few years, concerning pituitary gonadotrophins. While a number of recent find- ings are included in this discussion, the detailed presentation of new data has been avoided. The preparation in pure form of one of the gonadotrophins, the interstitial- tissue-stimulating hormone (ICSH), has been described. Great differences in the physicochemical properties of pure ICSH obtained from sheep and hog pituitaries have been demonstrated. Nevertheless the biological activity of the two compounds was found to be similar, both qualitatively and quanti- tatively. Similar though less pronounced chemical species differences have been noted also for sheep and beef lactogenic hormones. Species differences have thus become a factor to be reckoned with when comparing the properties of hormones obtained from different animal species. Biological differences be- tween follicle-stimulating hormones (FSH) from sheep and hog pituitaries may rest on chemical differences between hormones of the two sources. Besides these species differences, there are already well-established contrasts between gonadotrophins from the pituitary and those of chorionic (or endo- metrial) origin. It appears possible that the hormones secreted by the pituitary resemble more those of placental origin than those extracted from the gland. However, the small amounts of pituitary hormones circulating in the blood makes their purification as yet impracticable, and gland extracts have there- fore been generally used as source material for the isolation of hormones. The biological properties of FSH and ICSH prepared in this manner have been summarized. The intrinsic effects of ICSH which characterize it as a separate hormone have been considered separately from those "augmentation" effects which are evident only in the presence of FSH and may be simulated by non- specific agents which act merely by delaying the absorption of FSH. Fraenkel-Conrat, Li, and Simpson igi REFERENCES 1. Gurin, S.; Bachman, J. M., and Wilson, D. W.: Jl. Biol. Chem. 133:477, 1940. 2. Gurin, S.: Proc. Sec. Exper. Biol. & Med. 49:48, 1942. 3. Li, C. H.; Evans, H. M., and Wonder, D. H.: JI. Genrl. Physiol. 23:733, 1940. 4. Lundgien, H. P.; Gurin, S.; Bachman, J. M., and Wilson, D. W.: Jl. Biol. Chem. 142:367, 1942. 5. Fevold, H. L.; Hisaw, F. L., and Leonard, S. L.: .\mer. Jl. Physiol. 97:291, 1931. 6. Fevold, H. L.; Lee, M.; Hisaw, F. L., and Cohn, E. J.: Endocrinology 26:999, 1940. 7. Fraenkel-Conrat, H.; Simpson, M. E., and Evans, H. ^L: An. Fac. de med. de Montevideo 25:159. 1940- 8. Fraenkel-Conrat, H.; Li, C. H.; Simpson, \l. E., and Evans, H. M.: Endocrinology 27:793, 1940. 9. Fraenkel-Conrat, H.: Simpson, NL E., and Evans, H. M.: Proc. Soc. Exper. Biol. & Med. 45:627, 1940. 10. Li, C. H.; Simpson, M. E., and Evans, H. M.: Endocrinology 27:803, 1940. 1 1. Li, C. H.; Simpson, M. E., and Evans, H. M.: Science (n.s.) 92:355, 1940. 12. Li, C. H.; Simpson, M. E., and Evans, H. M.: Jl. Amer. Chem. Soc. 64:367, 1942. 13. Chow. B. F.; Greep, R. O., and Van Dyke, H. B.: Jl. Endocrinol. 1:440, 1939. 14. Greep, R. O.; Van Dyke, H. B., and Chow, B. F.: Jl. Biol. Chem. 133:289, 1940. 15. Shedlovsky, T.; Rothen, A.; Greep, R. O.; Van Dyke, H. B., and Chow, B. F.: Science (n.s.) 92:178, 1940. 16. Jensen, H.; Simpson, M. E.; Tolksdorf, S., and Evans, H. M.: Endocrinology 25:57, 1939. 17. Jensen, H.; Tolksdorf, S., and Bamman, F.: Jl. Biol. Chem. 135:791, 1940. 18. Fraenkel-Conrat, H.; Simpson, M. E., and Evans, H. M.: Endocrinology 27:809, 1940. 19. Evans, H. M.; Fraenkel-Conrat, H.; Simpson, M. E., and Li, C. H.: Science (n.s.) 89:249, 1939- 20. Greep, R. O.; Van Dyke, H. B., and Chow, B. F.: Anat. Recrd. 78:88, 1940. 21. Greep, R. O.; Van Dyke, H. B., and Chow, B. F.: Amer. Jl. Physiol. 133:303, 1941. 22. McShan, W. H., and Meyer, R. K.: Jl. Biol. Chem. 126:361, 1938. 23. McShan, \V. H., and Meyer, R. K.: Proc. Soc. Exper. Biol. & Med. 40:699, 1939. 24. Chen, G., and Van Dyke, H. B.: Proc. Soc. Exper. Biol. &: Med. 40:172, 1939. 25. Abramowitz, A. A., and Hisaw, F. L.: Endocrinology 25:623, 1939. 26. Li, C. H.; Simpson, M. E., and Evans, H. M.: Jl. Biol. Chem. 131:259, 1939. 27. Fraenkel-Conrat, H.; Simpson, M. E., and Evans, H. M.: Jl. Biol. Chem. 130:243, 1939. 28. Fraenkel-Conrat, H.; Simpson, M. E., and Evans, H. M.: Science (n.s.) 91:363, 1940. 29. Greep, R. O.; Van Dyke, H. B., and Chow, B. F.: Proc. Soc. Exper. Biol. & Med. 46:644, 1941- 30. Simpson, M. E.: Li, C. H., and Evans, H. M.: Endocrinology 30:969, 1942. 31. Fraenkel-Conrat, H.; Simpson, M. E.; Li, C. H., and Evans, H. M.: An. Fac. de med. de Montevideo 25:169, 1940. 32. Fevold, H. L.: Endocrinology 28:33, 1941. 33. Greep, R. O.: Proc. Soc. Exper. Biol. & Med. 44:214, 1940. 34. Fraenkel-Conrat, H.; Li, C. H.; Simpson, M. E., and Evans, H. M.: Proc. Soc. Exper. Biol. & Med. 48:723, 1941. ESTROGEN ASSAY IN THE HUMAN By S. C. FREED X From the DEPARTMENT OF ENDOCRINE GYNECOLOGY, LOYOLA UNIVERSITY MEDICAL school; DEPARTMENT OF MEDICINE, MICHAEL REESE HOSPITAL, CHICAGO, ILLINOIS ESTROGEN ASSAY IN THE HUMAN A NUMBER OF invcstigators have recently expressed the opinion that the . therapeutic efficiency of an estrogen can only be judged by testing its potency in the human. Most claims for the relative therapeutic effectiveness of estrogens have been made on the basis of their activity as determined in laboratory animals, principally the rat or mouse. The assumption that data so obtained can be accepted for the human has resulted in considerable con- fusion in the standardization of estrogen therapy. In the first place, results obtained from assays differ widely, as indicated by the fact that the rat unit of estrone as determined in different laboratories varies as much as several thousand per cent when compared to a weighed amount of crystalline mate- rial; the same holds true for assays in the mouse. The discrepancy in the assay of estrogens in rat or mouse cannot be entirely accounted for by technical differences in the performance of the assays, and it appears quite certain that each strain of rat or mouse has a different degree of sensitivity to any one estrogen. Similarly, comparisons of the potencies of different estrogens such as estrone and estradiol in the same laboratory cannot be judged as the true reflection of the relative therapeutic potencies of these substances, inasmuch as the ratios of activity as determined by different laboratories are far from constant. In a recent article a compilation of data on this subject reported by a number of investigators illustrates the inconsistencies in animal assays, lead- ing to the conclusion that any statement regarding the relative therapeutic activity of estrogens on the basis of animal assays is liable to considerable error and that assay in the human is at the present time the only hope for satisfactory therapetitic standards of estrogens.^ Several attempts have been made to assay the activity of estrogens in humans. Some investigators have utilized as an index of estrogen activity the changes in the vaginal mucosa of menopausal patients following estrogen administra- tion, much the same way as the castrate rodent is used.' When it is recognized, however, that untreated menopausal patients have varying degrees of prolif- eration of the vaginal mucosa, it does not appear that this would be a satis- factory means of assay. There is a lack of evidence that the vaginal epithelium of a group of menopausal women will respond to a definite amount of estrogen with a sufficient degi^ee of uniformity. Furthermore, the reading of vaginal smears in the human for assay purposes is liable to experimental error. An attempt has been made to utilize the changes in the menopausal endo- metrium following estrogen administration as a means of assay.' Such a tech- nique is not only cumbersome but also open to even severer criticism than that mentioned above for the use of vaginal smears, inasmuch as untreated meno- pausal patients may possess significant degrees of endometrial proliferation and, in fact, hyperplasia.' This factor would interfere greatly with assays based on endometrial changes. C195II 1 gS Estrogen Assay in the Human The author has selected as an end point in the assay of estrogens in humans the subjective response of menopausal patients. It is acknowledged that the evaluation of such a response may be obscured by numerous uncontrolled fac- tors. Nevertheless, this method has been selected for assay for a number of reasons, not the least of which is the fact that the chief purpose in administer- ing estrogens is to relieve the menopausal patient of her subjective symptoms. Such an assay requires no special technique and a large number of patients may be included in a study with little difficulty. In order to eliminate as many distracting factors from this study as possible, the author has established cer- tain criteria. Patients were selected who complained of moderate or severe menopausal symptoms; those who had complaints which were of doubtful origin or might possibly be confused wath psychic changes due to environ- mental or social complications were not included in the group tested. Patients were not included in the groups unless they had at least two to four hot flashes daily, together with other symptoms commonly found in the menopause, such as nervousness, irritability, and emotional instability. The psychic factors asso- ciated with any form of therapy involving subjective sensations were reduced to a miniinum by eliminating any therapeutic suggestion, such as a promise of beneficial results or leading questions concerning therapeutic responses. In addition, the subjective changes of all patients were evaluated by the au- thor alone in as constant a manner as possible, thus eliminating differences in interpretation of results which are prone to develop where a number of clinicians are treating the same group of patients. Furthermore, the estrogens were administered in several dosage levels in the manner which is used for assaying estrogens in laboratory animals. Two synthetic estrogens, hexestrol, and diethylstilbestrol, were thus assayed. Methods and Results Patients were given at different times one of three levels of the synthetic es- trogen hexestrol— 1 mg., 2.5 mg. and 5 mg. daily. Diethylstilbestrol was ad- ministered daily in dosages of 0.5 mg. and 1 mg. Therapy was usually started by administering the highest level of hexestrol. After three weeks of treatment the therapeutic response was evaluated and listed in terms of plus signs, 3+ being considered the optimal response, 2^ a satisfactory response, and 1^ a slight though definite response. In the evaluation of therapeutic response, the disappearance of hot flashes was used as one of the most important criteria of relief. Changes in nervousness, irritability, etc., were also considered in this evaluation. After the first period of treatment, the dosage was dropped to the next lower level and maintained in this manner for three weeks or more. After the patient's response w'as evaluated at the medium level, the lowest level was then administered for three or more weeks and further response noted. The patients were then given the second estrogen, diethylstilbestrol, and their clinical course with two dosage levels followed in a similar manner. A number of patients received medication in the reverse order, first diethvlstilbestrol, S. C. Freed 197 then hexestrol. Some patients were unable to undertake the entire course of assay due to such reasons as faikne to follow directions satisfactorily and re- fusal to continue therapy because of unpleasant symptoms which developed. The following table contains the results of assay of both estrogens by the above-described method, together with the incidence of unpleasant symptoms. TABLE 1 Therapeutic Response of Menopausal Patients following Administration of Hexestrol AND DiETHYLSTILBESTROL BY MoUTH Hexestrol Diethylstilbestrol Dailv dose I mg. 2.5 mp. 5 mg. o.s mg. I mg. No. of patients Response No. of patients Response No. of patients Response No. of patients Response No. of patients Response 21 10 6 neg. + + + 7 13 29 II neg. + + + + 15 5 ^3 neg. + ++ +++ 9 17 18 9 neg. + + + + 4- + 5 10 27 23 neg. + + + + + + 37 60 65 53 65 Incidence of toxic reactions • NHD, nausea with or without dizziness and headache; N, nausea, H, headache; D, dizziness. A number of patients were questioned at certain times as to which of the two preparations they preferred. The results of this questionnaire are as follows: TABLE 2 28 found 2.5 mg. daily hexestrol equal to 0.5 mg. daily diethylstilbestrol 14 found 2.5 mg. daily hexestrol better than 0.5 mg. daily diethylstilbestrol 4 found 2.5 mg. dailv hexestrol worse than 0.5 mg. daily diethylstilbestrol 34 found 5 mg. daily hexestrol equal to i mg. daily diethylstilbestrol 6 found 5 mg. daily hexestrol better than i mg. daily diethylstilbestrol 8 found 5 mg. daily hexestrol worse than i mg. daily diethylstilbestrol It is apparent from the data contained in table 1 that a satisfactory thera- peutic dose of diethylstilbestrol is 0.5-1.0 mg. daily, and an equivalent thera- peutic dose of hexestrol is 2.5-5.0 mg. daily, the minimal effective therapeutic doses being 0.5 mg. diethylstilbestrol and 2.5 mg. hexestrol daily. It will be noted that there was a larger number of patients who failed to respond to 5 mg. daily of hexestrol than to 2.5 mg. daily of the same substance. This 1 gS Estrogen Assay in the Human apparent paradox can be explained by the fact that the patients were first administered the higher level. Those who failed to respond did not receive the 2.5 mg. daily dose. The seven patients who failed to respond to the 2.5 mg. daily dose had received some beneficial results when they were given 5 mg. daily. In regard to the toxic manifestations of this therapy, which is of considerable interest inasmuch as the unpleasant symptoms following oral administration of these estrogens are considered to be their greatest disadvantage, it appears that hexestrol is significantly less toxic than diethylstilbestrol, although the incidence of toxic reactions from hexestrol administration is appreciable. Ad- ditional data seem indicated on this phase of the study before definite con- clusions can be drawn as to the relative toxicity of these estrogens. Discussion By the use of different dosage levels, a satisfactory evaluation of the therapeutic potency of estrogens can be obtained by using the relief of menopausal symp- toms as an end point. Nonspecific factors are minimized by the proper selec- tion and handling of patients. With the use of the technique described above, there is no need for control with either untreated patients or patients who receive placebo medication inasmuch as each dosage level controls the next. The psychic factor of administering some form of medication is thus reduced. In this regard, the author has observed a number of patients who responded equally well on all levels of estrogens and who when placed on placebo medica- tion also responded satisfactorily. The data obtained from these were not in- cluded in the tables since the inclusion of this type of patient in a group for study renders results less significant statistically. With the use of multiple dosage levels it is not necessary for statistical purposes to treat a large series of patients receiving a single dosage since, as can be noted from the two tables, these is some cancellation of experimental error at the different levels. Thus, in table 1, 2.5 mg. hexestrol seems somewhat superior to 0.5 mg. hexestrol, but 5 mg. hexestrol is somewhat inferior to 1 mg. diethylstilbestrol. It is also expected that any new estrogen can be easily compared in potency to estrogens which have already been tested by this assay method. Furthermore, the mini- mal therapeutic dose can readily be determined by the use of multiple dosage levels. The results of this study are not in accord with those of Bishop and co- workers^ regarding the relative therapeutic activity of hexestrol and diethyl- stilbestrol. These workers report that these compounds are almost equal in potency. My results indicate that hexestrol is one-fifth as effective as diethylstil- bestrol. In addition, they claim hexestrol to be relatively free of untoward reactions. In my hands hexestrol appears to be less toxic than diethylstilbes- trol in equivalent therapeutic doses, although the incidence of unpleasant symptoms is sufficiently high to warrant caution in its use. S. C. Freed igg Conclusions Two synthetic estrogens, diethylstilbestrol and hexestrol, have been tested for their therapeutic efficiency by assay in humans. The subjective response of menopausal patients Avas used as the end point in this assay and the usual uncontrolled factors encountered in such a study were eliminated to a great extent by using multiple dosage levels of estrogens, as well as other precau- tions. From the results obtained, it can be concluded that hexestrol is about one-fifth as effective as diethylstilbestrol in the human. It is indicated that hexestrol is significantly less liable to induce undesirable reactions than di- ethylstilbestrol at equivalent therapeutic dosages. REFERENCES 1. Freed, S. C: Jl. Amer. Med. Assn. 117:1175, 1941. 2. Papanicolaou, G. N., and Shorr, E.: Amer. Jl. Obstet. & Gynecol. 31:806, 1936. 3. Werner, A. A.: Internatn. Abstr. Surg. 73:49, in Surg. Gynecol. & Obstet., 1941. 4. Novak, E., and Richardson, E. H.: Amer. Jl. Obstet. & Gynecol. 42:564, 1941. 5. Bishop, P. M. F.; Bowes, R. K.; Boycott, M.; Kellar, R.; MacGregor, T. M., and Murless, B. C.: Lancet 238:629, 1940. FUNCTIONAL INTERRELATION OF CEREBRAL CORTEX WITH BASAL GANGLIA AND CEREBELLUM By J. F. FULTON, M.D. X From the LABORATORY OF PHYSIOLOGY YALE UNIVERSITY SCHOOL OF MEDICINE NEW HAVEN, CONNECTICUT FUNCTIONAL INTERRELATION OF CEREBRAL CORTEX WITH BASAL GANGLIA AND CEREBELLUM* HAR\ Ev GUSHING oncc good-natuiedly accused Herbert Evans of being "pi- tuitary minded," and I believe that there was a prompt rejoinder from Doctor Evans to the effect that Gushing liad lately transferred his "minded- ness" as well as his affections to the hypothalamus. Had it been a three-cor- nered exchange, one might have suggested that these two seats of Cartesian turmoil are connected by silken threads of functional integration. In my mind's eye I can see Herbert racing around the sella turcica, peeping over the posterior clinoid or looking menacingly up through the diaphragm— only to meet the penetrating gaze of a diminutive Harvey Gushing seated securely on the floor of the third ventricle and exclaiming: "What, Herbert, more frac- tions? Stop it— I'll stir up one of them from here, and create a case of diabetes that even Jack Peters cannot control." What other excuse can I offer for presenting the nervous system to Herbert Evans on his birthday? I fear he may not like it— but even so, he surely will be pleased by the thought that he and Gushing are still playing hide-and-seek with one another above and around the sella. And what game on the part of two men ever proved more richly rewarding to us all? In selecting for discussion the broad theme of the physiological relation be- tween the two great extrapyramidal systems, (i) the cortico-strio-nigral, and (ii) the cortico-ponto-cerebellar, I am aware that I am taking liberties with one of the classical concepts of clinical neurology, for in common parlance the "extrapyramidal" system begins with the basal ganglia and ends some- where in the spinal cord through the reticulo- and rubro-spinal projections. From the physiological standpoint, this is an indefensible concept, since in higher animals all subcortical motor nuclei, including the cerebellum, are under the direct control of the cerebral cortex, and it is impossible in a func- tional analysis to divorce the subcortical from the cortical centers of motor integration. The Gortico-strio-nigral System One may take for granted a knowledge of the principal subcortical nuclei of the extrapyramidal system: the neostriatum (caudate and putamen) discharg- ing through the paleostriatum or globus pallidus, which in turn discharges by the ansa lenticularis to substantia nigra and other tegmental nuclei. In early embryological history, the paleostriatum and its motor projections were the principal motor pathways from the forebrain, but with the development of the cerebral cortex, the striate bodies became dominated by direct and in- * Modified from an address to the Philadelphia Neurological Society, read March 28, 1941. 1:203] 204 Functional Interrelations o£ the Brain Stem direct projections from the cerebral cortex; indeed, they have become so intimately linked with the motor functions of the higher cortical level that in man and higher apes it is probably impossible for them to function in the organization of motor acts without the cerebral cortex itself. The interrelation between basal ganglia and the cortex has indeed become so close that they appear to function as a homogeneous unit. The basis for these generalizations comes from three lines of evidence: anatomical, electrical, and from studies involving seriatim ablations of the cerebral cortex and the striatal nuclei. Anatomical Studies.— The anatomical evidence of interconnections between cortex and basal ganglia is less impressive than the physiological. Nearly all observers working with the Marchi technique, after ablation of areas 4 and 6 of the cerebral cortex, have found degenerating fibers passing into the caudate and putamen, and some also into the globus pallidus. But with Marchi sec- tions there is no absolute proof that the fibers actually end in the area in which they are seen to disappear (Levin,^ Mettler^. A carefully controlled study by Verhaart and Kennard,* in which individual areas were destroyed by the thermocoagulation technique of Dusser de Barenne, indicated that fibers from areas 4 and 6, and especially from the intermediate-strip area 4-s of Marion Hines, passed into the striatum; but as with the earlier studies, there was no certain proof that they terminated there. Ramon y Cajal* had described corticostriatal fibers with silver preparation and he has also observed dichoto- mizing fibers from corticospinal projections passing into the putamen at the level of the internal capsule. These, however, were fine fibers which one could not expect to follow by Marchi degeneration. Electrical Evidence.— T>usser de Barenne and McCulloch^ found that strych- ninization of certain areas in the macaque's precentral cortex evoked con- spicuous electrical activity in the striate nuclei. The observation was one of challenging interest because of associated effects elsewhere in the cerebral cortex, and because of the sharply circumscribed character of the response in the striatum itself. The precentral convolution is divided into three pri- mary areas: area 4, area 6, and an intermediate-strip region, first defined by Marion Hines"' which has been designated "area 4-s."* When A4-S is strych- ninized, spontaneous electrical activity in A4 disappears (see fig. 1) and its threshold for stimulation rises. Dusser de Barenne and McCulloch pointed out that this suppression of electrical activity by local strychninization is not mediated transcortically, but by a complex circuit from A4-S to the caudate nucleus, thence to the thalamus and from there back to area 4. The caudate nucleus thus becomes specifically activated by stimulation of area 4-s; they found furthermore that strychninization of areas 4 or 6 had no such effect on the caudate. Isolated destruction of the caudate, and of no other part of the basal ganglia, caused disappearance of the strychnine suppression from area 4-s. * To distinguish arm, leg, and face, it has become customary to refer to the major sub- divisions as L4, A4, F4, L4-S, etc., designating the leg, arm, and face regions respectively. John F. Fulton 205 Later studies by Dusser de Barenne, Garol and McCulloch,^ reported at the meeting of the Association for Research in Nervous and Mental Disease in December, 1940, indicate that there are other suppressor areas, notably areas 8-s and 2-s (fig. 1). On stimulation, both regions may cause activation of the caudate nucleus in a sharply circumscribed area, 8-s activating the anterior Fig. 1. A diagram of the cerebral cortex of the common Rhesus monkey {Macaca mulatto.) showing the principal suppressor areas (8-s, 4-s. 2-s, 19-s) together with the primary motor regions 4L (leg), 4A (arm), 4F (face). (From Dusser de Barenne, Garol and McCulloch: Jl. Neurophysiol. 4:325, 1941) part of the caudate, 4-s the medial part, and 2-s the tail of the caudate. Thus all of the suppressor areas of the cortex converge in a spatially oriented manner upon the caudate nucleus. With regard to the other parts of the basal ganglia, it is no doubt significant that, whereas areas 4 and 6 fail to project to the caudate, they do, when strych- ninized, cause strong electrical activation of the putamen, and area 6 also affects the external segment of the globus pallidus. Although the suppressor regions exhibit an anteroposterior organization of their projections, there is less precise localization within the putamen and globus pallidus. To quote Dusser de Barenne, Garol and McCulloch: From a purely anatomical standpoint, that spatially separated areas 4 and 6 should project to one structure, the putamen, while 4-s and 8-s project to another, the nucleus caudatus, is truly surprising, for 4-s lies between 4 and 6, and 6 lies between 4-s and 8-s. On the other hand, when one realizes that 4 and 6 are alike in giving motor responses and that 4-s and 8-s are alike in giving suppression, it becomes highly probably that the projections from 4 and 6 may converge somewhere else, and it ceases to be so surprising to discover these discrete con- vergences in the corpus striatiun. 2o6 Functional Interrelations o£ the Brain Stem There is also impressive evidence, based upon electrical reactiveness, point- ing to a close functional interrelation between the cerebral cortex and the basal ganglia. Removal of the cerebral cortex, leaving the basal ganglia intact, does not preserve the electrical excitability of these structures, save for certain doubtful reactions in the lower vertebrate forms (Rioch and Brenner^). In monkeys and chimpanzees, weeks or months after the cerebral cortex has been removed the basal ganglia are quite unstimulable as far as somatic reactions are concerned (Fulton^"). The striking studies of Mettler, Ades, Lipman and Culler," on the other hand, in which the basal ganglia have been stimulated during concurrent stimulation of an intact cerebral cortex, indicate that from certain striatal regions suppression of cortical and motor activity may be brought about; during spontaneous phasic movements originating in the cerebral cortex, stimulation of the caudate causes suppression of these reactions— a true central inhibition. Such suppression might have been anticipated from the studies of Dusser de Barenne and McCulloch. Seriatim Lesions of Cortex and Basal Ganglia.— The consequences of isolated regional ablations of the motor area are well known in monkeys, chimpanzees, and man. Two generalizations can be made concerning the effects of such lesions. The first is that a lesion restricted to area 4 causes motor weakness accompanied by flaccidity of the affected extremities; transient spasticity of the digits may develop in three to four weeks after an isolated area 4-lesion (Denny-Brown), but the predominant feature of corticospinal interruption is a flail paresis. If the corticospinal projections are interrupted at the level of the medullary pyramids, an enduring and purely flaccid paralysis results (Tower^") without spasticity at any joint at any time. This observation entirely harmonizes with the results of area 4-ablation. With area 4-ablations, some extrapyramidal projections are inevitably interrupted, and these are no doubt responsible for the transient spasticity that appears in the digits after an area 4-lesion. The second generalization from regional ablation of the cerebral cortex is that hypertonicity and spasticity (that is, generalized increase of resistance to passive manipulation) result only when extrapyramidal projections are inter- rupted, either at the cortical level or, as we shall see later, at subcortical regions. When lesions of the caudate, putamen, and globus pallidus are added to specific ablations of the cerebral cortex in adult animals— either when the striatal lesion is made simultaneously or at an interval after the cortical abla- tion—the motor deficit is invariably greater than with isolated removal of a given cortical area. Primary lesions of the caudate or putamen, without injury to the cortex, cause inconspicuous motor deficit. Only when combined with lesions of the cortex is the deficit grave. Primary lesions of the globus pallidus, however, cause moderate deficit in adult animals, and if the pallidal lesions are bilateral the effects are more severe and lasting. Such animals exhibit bilateral paresis and, although able John F. Fulton 207 to climb, walk, and run, they have obvious difficulty in finer prehension move- ments. Spontaneous movements, moreover, are considerably slowed. Resist- ance to passive manipulation is increased and marked tremor develops, not seen in unilateral preparations. The tremor is definitely augmented by voli- tional effort and as such possibly differs from Parkinson's tremor. This I shall return to in a moment. Further and highly impressive evidence of the intimate association between the cerebral cortex and the basal ganglia has come from Kennard's studies" of the effect of ablating the cortex and basal ganglia of infant monkeys. Lesions of the motor and premotor areas, if made shortly after birth, cause no serious impairment in motor movement. At birth, motor acts are integrated largely at the subcortical level. Lesions of the basal ganglia, on the other hand, if made in infancy cause grave impairment of the motor behavior patterns, indicating that at this stage in the animal's development the cortex has not yet assumed control of basal ganglia activity. With the development of iso- lated movements, as Marion Hines" and others have emphasized, the cerebral cortex assumes dominance in the regulation of motor patterns. As the animal matures, the basal ganglia no doubt remain in the background, giving sup- port for complex adjtistments; all available evidence indicates that they func- tion more in the postural sphere than in that of phasic movement. Biu the two spheres of activity are so closely related that adequate separation is difficult, for, as Stanley Cobb once said: "Every phasic movement begins and ends in a posture." Tremor and the Clinical Basal-Ganglia Sy7idromes.—M.?iny attempts have been made in the past to reproduce the involuntary movements which form such a conspicuous part of the so-called basal-ganglia syndrome in men, that is, Parkinson tremor, chorea, and athetoid movements. Kennard^" has been able to induce in monkeys a tremor which often continues during rest and is greatly exaggerated by motion. This has been done by two different pro- cedures: (i) bilateral ablation of the globus pallidus, unassociated with lesions of the cerebral cortex, and (ii) bilateral lesions of caudate and putamen ac- companied by subtotal removal of the precentral convolution (either area 4 or area 6). Tremor established in this manner can be abolished if the remaining portion of the precentral convolution is removed. Integrity, therefore, of the motor projections, extrapyramidal or pyramidal, from the cortex is essential to the production of tremor associated with lesions of either the neostriatum or the paleostriatum. This observation has a bearing upon the neurosurgical procedures recently proposed for the relief of Parkinson tremor, chorea, hemi- ballismus, etc., and will be discussed below. Kennard has had less success in producing athetosis and chorea. Athetoid movements have been seen in several infant monkeys following bilateral lesions of the caudate and putamen. Such movements, although unmistakable when present, are transient and difficult to reproduce from one animal to the next. A characteristic jerking chorea was induced in one chimpanzee by a 2o8 Functional Interrelations o£ the Brain Stem lesion involving area 6 and the head of the caudate nucleus. It persisted for several weeks, then abruptly diminished in intensity, but was occasionally seen during a period of five or six months, whenever the animal became ex- cited. We have been unable in two subsequent chimpanzees, with a similar lesion, to reinduce the chorea, and we are as yet unable to state what the es- sential lesion may have been. In the analysis of the tremors and involuntary movement of our striatal animals, we have been frequently aware of certain similarities between the tremors of corticostriatal origin and those of corticocerebellar origin, and now I would like briefly to describe a series of recent studies on the physiology of the cerebellum which have an intimate bearing upon the problem in hand. The Cortico-ponto-cerebellar System From the cerebral cortex also arises a vast system of fibers which converge on the pontine nuclei, and from there pass their impulses to the cerebellum. The corticospinal projections, in addition, are said to send branching fibers to the pontine nuclei which directly influence cerebellar mechanisms. The cerebellar hemispheres have developed in evolutionary history pari passu with the elabo- ration of the cerebral cortex, and all authorities are agreed that the two organs have a close functional interrelation. It is customary to speak of the cortico- strio-nigral system as being primarily concerned with postural reactions, whereas the cortico-ponto-cerebellar system is believed to be associated pri- marily with phasic movement. Neither generalization, however, can bear too close scrutiny. The cerebellum has three primary divisions, two arising from the ancient brain— the posterior lobe and the anterior lobe, both midline structures— and the third, the neocerebellum, comprising the cerebellar hemispheres, which are intimately associated with the hemispheres of the cerebrum. Posterior Lobe.— This includes two primary divisions: (a) a spinocerebellar part, consisting of pyramis and uvula, and (b) the flocculonodular lobe, which is primarily vestibular in its connections. Dow^*" has shown that ablation of the flocculonodular lobe causes disturbances of balance similar to those seen in grosser form when the vestibular nuclei are injured, and it is obvious that this ancient part of the cerebellum plays an important role in maintaining the equilibrium of the body. Less is known concerning the pyramis and uvula, except that their ablation in conjunction with that of the flocculonodular lobe considerably exaggerates the disturbance of equilibrium caused by ablation of the flocculonodular lobe alone. Their isolated ablation, however, does not seriously affect equilibrium. Anterior Lobe.— The anterior lobe, also called paleocerebellum, is separated from the neocerebellum by the fissura prima and is made up of three divisions: the culmen, centralis, and lingula. Stimulation of the anterior lobe, as origin- ally shown by Miller and Banting," causes inhibition of decerebrate rigidity, and, in a lightly anesthetized animal with cortex intact, inhibition of extensor John F. Fulton 209 postures. Until very recently, few attempts had been made to discover whether the responsiveness of the anterior lobe is discrete, that is, whether there is separate representation of upper and lower extremities. All are agreed, how- ever, that the homolateral side of the anterior lobe influences the homolateral extremities to a greater extent than the contralateral. The recent studies of Gervase Connor^^ indicate that, both in dogs and in monkeys, a maximal extensor release occurs when the anterior lobe is removed as an isolated entity. The positive supporting reactions of Magnus become grossly exaggerated, with head retraction and marked opisthotonos; in a pains- taking analysis Connor finds that this extensor hyperactivity persists after de- afferentation of the neck muscles, after bilateral labyrinthectomy and after renroval of both cerebral hemispheres, but is destroyed by a posterior root section which interrupts the proprioceptives to the extremity under exami- nation. The exaggeration of positive supporting reaction is so intense, follow- ing anterior lobe lesions, that it has given opportunity to study the question of functional localization within the anterior lobe. Doctor Connor permits me to mention the following new disclosures. A. Ciilmen.— When the culmen is removed as an isolated entity, leaving centralis and lingula intact, positive supporting reactions are foiuid only in the hindlimbs, the forelimbs being quote unaffected. If one lateral half of the culmen is removed, positive supporting reactions are seen only in the hindlimb on the same side. From this, Connor concludes that ftmctional representation of the hindlimbs can be demonstrated in the culmen and in the culmen alone. B. Centralis.— Isolated removal of the posterior centralis causes positive sup- porting reactions elicitable only in the forelimbs, and when the lateral half of the centralis is removed the reactions are restricted to the forelimb on the side of the lesion. From this Connor concltides that the posterior centralis pre- sides over the antigravity reactions of the upper extremity. C. Li?igi{ la. —The lingula is peculiar in being the only portion of the ante- rior lobe which receives vestibular in addition to spinocerebellar fibers. \V'hen the entire anterior lobe is removed one of the striking symptoms, in addition to positive supporting reactions of the extremities, is an extreme head retrac- tion with obvious incoordination of the neck musculature. After isolated lesions of culmen and centralis, this symptom does not occur and is only found when the lingula is included in the ablation. Owing to its anatomical position, it has been impossible as yet for Connor to remove the lingula as an isolated entity, but from the evidence just presented it would seem an entirely logical conclusion that the lingula presides over the neck musculature, and with its vestibular connection stands as an integrator of the vestibtilar and the tonic neck-reflex mechanisms. You will no doubt ask, how are these mechanisms related with the cerebral • cortex? Their relation is less intimate than with the neocerebellum, btit in Connor's analysis of the syndrome of the culmen, centralis, and lingula re- spectively he finds that all reactions, such as the positive supporting reflex, are 210 Functional Interrelations of the Brain Stem still further released -when the cerebral cortex is removed. Connor has not yet had opportunity to study the effects of isolated cortical ablations in relation to isolated ablations of the anterior lobe; but it is clear that the postural mechanisms originating in the cerebral cortex interact with those of the an- terior lobe. Neocerebelhim.— It is less essential to dwell upon cortico-neocerebellar in- terrelations since they are more commonly recognized. Phasic movements are grossly disturbed following removal of the neocerebellar cortex. When large lesions of the neocerebellum occur, errors of rate, range, force, and direction of volitional movements commonly take place, and the discontinuities of move- ment which constitute the summation of these errors is generally referred to as "cerebellar tremor." One thing frequently lost sight of in the analysis of cerebellar syndromes is that cerebellar tremor does not arise in the cerebellum, for it develops after its complete removal. It is due to action in some other part of the nervous system, and one can readily prove that cerebellar tremor, so-called, results from activity of the cerebral cortex, since the discontinuities of movement characteristic of cerebellar disease completely disappear on removal of the cerebral cortex (Fulton, Liddell and Rioch"") or, in monkeys, on bilateral abla- tion of areas 4 and 6. Hence, cerebellar tremors and striatal tremors have this in common, namely, that they depend upon the integrity of the motor projections from the cerebral cortex; and the similarities between the two types of tremor are greater than is sometimes appreciated— or perhaps one should say, the differences are less than our textbooks would lead us to suppose. Cerebellar tremors are most con- spicuous at the end of a volitional movement and are sometimes referred to as "terminal tremors." When a monkey with a neocerebellar lesion reaches for a banana, the tremor, scarcely obvious at first, becomes grossly exaggerated just as the animal reaches its objective. A Parkinson tremor, on the other hand, tends to be more exaggerated at the beginning of movement and may be ironed out by the time the movement is completed. I vigorously object to the distinction commonly made between striatal and cerebellar tremors, namely, that one is a tremor of rest and the other a tremor of action. The Parkinson tremor completely disappears during sleep and is present only when the subject is attempting by voluntary effort to maintain the body in a certain posture. It is in reality a tremor resulting from volitional control over postural mechanisms, whereas cerebellar tremor appears as a re- sult of volitional control of phasic movements. Both, basically, are tremors of action. Summary and Conclusions This survey of the two divisions of the extrapyramidal system leads to several general conclusions, as well as to a number of practical considerations which may be summarized as follows: John F. Fulton 2 1 1 1 . Phasic movement is integrated by that part of the cortico-ponto-cerebellar system which passes to the neocerebellum. Destruction of the neocerebellum leads to tremors of action which are most conspicuous at the termination of volitional movements. Such action tremors disappear when the precentral con- volution of the cortex is removed. 2. Postural adjustments are integrated by both systems, the cortico-strio- nigral and that part of the cortico-ponto-cerebellar which joins with the paleo- cerebelltmi.The antigravity and equilibritory postures are adjusted by cortical mechanisms acting in association with the anterior and posterior lobes of the cerebellum, whereas the other more complex postural adjustments, not having primarily to do with gravity or the position of the body in space, find integra- tion through the cortico-strio-nigral system. The cerebral cortex thus becomes the common focus of integration of the ex- trapyramidal systems both of the striatum and of the cerebellum, and through thus coordinating the postural with the phasic mechanisms absolute smooth- ness and precision of action are made possible. "Posture," it is said, "follows movement like a shadow," and this would be wholly impossible if the two were not integrated from a common level. Other more practical considerations emerge from this concept of interaction of the postural and phasic mechanisms, namely, the physiological basis for surgical intervention in the cortico-striatal syndromes, and possibly, also, in the cortico-cerebellar. The -work of Bucy and Buchanan^ (see also Bucy"^, Putnam,^ Meyers,^* and particularly that of Klemme,^ has focused attention upon the possibility of relieving patients afflicted with severe Parkinsonism, choreo-athetosis, hemi- ballismus, etc., through a regional cortical ablation. Similarly, Aring and Ful- ton"" showed experimentally that severe cerebellar tremors could be abolished by precentral lesions, but, as far as I am aware, no one has attempted to apply this clinically— possibly for very good reasons, namely that the motor deficit would be worse than the tremor. In the case of severe hemichorea, and in certain cases of recalcitrant Parkin- son tremors, conspicuous and apparently enduring relief has followed ablation of area 4 or of area 6 from the cerebral cortex on the side opposite the affected extremity. Klemme's operation necessitated making a large lesion in the pre- motor region, leaving area 4 essentially intact. Bucy, on the other hand, has reported"^ several excellent results from ablation of area 4 (involving only the posterior few millimeters of area 6). It is not possible at present to say which is the operation of choice. I feel certain that complete ablation of areas 4 and 6 will arrest a Parkinson tremor, a hemiathetosis or hemichorea, as well as severe cerebellar tremor. Area 4-ablation will produce much the same result, as far as cerebellar tremor is concerned, but there tends, with recovery of motor power, to be a redevelopment of the cerebellar symptoms. This has likewise proved true in certain clinical cases of hemichorea in which the symptoms have ultimately returned following small precentral lesions. From the physio- 2 1 2 Functional Interrelations o£ the Brain Stem logical standpoint, an area 4-ablation is likely to be more effective than an ablation of area 6; but if, as Klemme maintains, a large ablation of area 6 gives symptomatic arrest for an indefinite period it would clearly be the operation of choice, since area 6-lesions are accompanied by far less ultimate motor deficit than in the case of lesions of area 4. This field of work is still in its experimental stages and I believe that the solution can come only through further clinical experience. I am confident, furthermore, that in severe cases the operation is entirely justifiable and I sincerely hope that physiologically minded surgeons and neurologists will continue to encourage studies of this character, for they may bring relief in conditions which in years past have been regarded as entirely hopeless. REFERENCES 1. Levin, P. M.: J. comp. Neurol. 63:369, 1936. 2. Mettler, F. A.: J. comp. Neurol. 61:221, 509; 62:263; 63:25, 1935. 3. Verhaart, W. J. C, and Kennard, M. A.: J. Anat. 74:239, 1940. 4. Ramon y Cajal, S.: Histologic dii Systeme Nerveux de I'Homme et des Vertebres (Paris: 1909-1911). 5. Dusser de Barenne, J. G., and McCulloch, W. S.: J. Neurophysiol. 2:319, 1939. 6. Hines, M.: Amer. J. Physiol. 116:76, 1936. 7. Hines, M.: Johns Hopk. Hosp. Bull. 60:313, 1937. 8. Dusser de Barenne, J. G.: Garol, H. W., and McCulloch, W. S.: Res. Publ. Ass. nerv. ment. Dis. (1940) 21:246, 1942. 9. Rioch, D. M., and Brenner, C: J. comp. Neurol. 68:491, 1938. 10. Fulton, J. F.: Trans. Coll. Phys. Philad. 8:157, 1940. 1 1. Mettler, F. A.; Ades, H. W.; Lipman, E., and Culler, E. A.: Arch. Neurol. Psychiat. 41:984, 1939- 12. Tower, S. S.: Brain 63:36, 1940. 13. Kennard, M. A.: Arch. Neurol. Psychiat. 48:227, 1942. 14. Hines, M., and Boynton, E. P.: Contr. Embryol. Carn. Instn. 28:309, 1940. 15. Kennard, M. A., and Fulton, J. F.: Res. Publ. Ass. nerv. ment. Dis. (1940) 21:228, 1942. 16. Dow, R. S.: Arch. Neurol. Psychiat. 40:500, 193S. 17. Dow, R. S.: J. comp. Neurol. 68:297, 1938- 18. Miller, F. R., and Banting, F. G.: Brain 45: 104, 1922. 19. Connor, G. J., and German, W. J.: Trans. Amer. Neurol. Ass. 1941:181. 20. Fulton, J. F.; Liddell, E. G. T., and Rioch, D. M.: Arch. Neurol. Psychiat. 28:542, 1932. 21. Bucy, P. C, and Buchanan, D. N.: Brain 55:479, 1932. 22. Bucy, P. C: Res. Publ. Ass. nerv. ment. Dis. (1940) 21:551, 1942. 23. Putnam, T. J.: Res. Publ. Ass. nerv. ment. Dis. (1940) 21:666, 1942. 24. Meyers, P.: Res. Publ. Ass. nerv. ment. Dis. (1940) 21:602, 1942. 25. Klemme, R. M.: Res. Publ. Ass. nerv. ment. Dis. (1940) 21:596, 1942. 26. Aring. C. D., and Fulton, J. F.: Arch. Neurol. Psychiat. 35:439, 1936. THE SOLUBILITY OF PROTEINS AND THEIR SEPARATION FROM MIXTURES WITH SPECIAL REFERENCE TO SERUM By ARDA ALDEN GREEN From the DEPARTMENT OF PHARMACOLOGY WASHINGTON UNIVERSITY SCHOOL OF MEDICINE ST. LOUIS, MISSOURI THE SOLUBILITY OF PROTEINS AND THEIR SEPARATION FROM MIXTURES WITH SPECIAL REFERENCE TO SERUM THE PROCEDURES Ordinarily used for the isolation of individual proteins from mixtures such are are found in plasma or in extracts of muscle, placenta, liver, kidney, or endocrine organs are relatively simple. Isolation depends upon solubility in a given medium under given conditions, and the process of finding the optimum conditions is too often likely to be a matter of art or of luck. The physicochemical characteristics of the molecule undoubtedly determine its solubility but the protein must be isolated before such charac- teristics can be determined. There are, however, certain general, more or less theoretical considerations governing the solubility of proteins which will be summarized here briefly. In general, solubility depends upon the type of solvent, the presence of electrolytes, the kind and concentration of such electrolytes, the pH, the tem- perature, and the presence of other organic materials, especially other proteins. These factors must all be considered in the extraction as well as in the puri- fication of proteins. To avoid conftision it may be well to define the ordinary classes of protein. Albumins are water-soluble proteins precipitated only in relatively high con- centrations of neutral salts. Globulins are precipitated in lower concentrations of neutral salts and may be further divided into (i) water-insoluble globulins, true globulins or etiglobulins, and (2) water-soluble or pseudoglobulins. The original definition of a globulin as a protein insoluble in water and soluble in low concentrations of salts is still applicable. The fraction of serum precipita- ble in one-third saturation -with ammonium sulfate can no longer be termed "euglobulin" since it contains much water-soluble globulin and since water- insoluble globulins are found also in the "albumin" fraction. Separation by Isoelectric Precipitation Separation of proteins from aqueous solutions of low ionic strength depends upon isoelectric precipitation. Etiglobulins exhibit minimum solubility in the neighborhood of the isoelectric point and increase in solubility at either a lower or a higher pH, where the molecule has a higher positive or nega- tive charge. Horse hemoglobin, serum euglobulins, fibrinogen, prothrombin, pneumococcus and meningococcus antibodies, the coagulant from the pla- centa or limg, myosin from muscle, and certain hormones from the anterior pituitary may all be purified to varying degrees by this means. If more than one euglobulin is present in a solution, advantage may be taken of differences in isoelectric points or differences in solubility in neutral salt solutions. It should be noted that some proteins are more nearly insoluble than others at 2l6 Solubility o£ Proteins their isoelectric points, that small changes in electrolyte concentration may have a great effect on the pH of minimum solubility, and that solubility may be greatly influenced by temperature. Solubility in Concentrated Salt Solutions In concentrated solutions of neutral salts proteins are precipitated. The solu- bility of pure proteins may be described by the linear equation log ^ = iS - KJtx (i) where 5 is the solubility in grams per liter, ^i is the ionic strength per 1,000 gm. 1.0 -J o (/I u. 0 < a o -I T.0 Mf1,1^0. a rijsq, A Kj-ipq, \ \. 0 2 4 6 8 IONIC STRENGTH Fig. 1. The solubility of carboxyhemoglobin at 25° and pH 6.6 in concentrated solutions of various electrolytes. (From Green, A. A.: Jl. Biol. Chem. 93:512, 1931.) water, ^ the intercept constant, and K/ the "salting out" constant. If the logarithm of the solubility be plotted against the concentration of salt or against the ionic strength, a straight line results. If the quantities are expressed in terms of units per liter or in terms of mol fractions, the straight line still holds but the values of the constants are different. This form of equation was first applied to proteins by Cohn^ using the data of S0rensen and H0yrup^ and of Chick and Martin^ for the solubility of egg albumin in ammonium sulfate, and of S0rensen* for the solubility of pseudoglobulin in ammonium sulfate. a. Effect, of Type of Electrolyte on Solubility Uni-univalent salts are relatively ineffective in precipitating proteins. Only the more insoluble proteins such as fibrinogen are precipitated by these salts even when they are used in high concentrations. Higher valent salts produce much higher ionic strengths and are much more effective as precipitating agents. Arda Alden Green 217 Hemoglobin is a globulin. Its solubility is increased upon the addition of low concentrations of salt^ and it is "salted out" at higher concentrations of salt. Figure 1 represents the solubility of crystalline horse carboxyhemoglobin' at constant pH and temperature in concentrated solutions of various electro- lytes. The slopes of the lines, and therefore the K/ values vary with the kind of salt used. For this protein K/ decreases in the order, KH,PO^ + K2HPO , Na,SO„ Na^CgH-O,, (NH,),SO„ and MgSO,. This is essentially the same A.OJU ZD 3.Q JU. 4.0 Fig. 2. The solubility of hemoglobin in concentrated phosphate buffers. of varying temperature and pH. (From Green, A. A.: Jl. Biol. Chem. 93:507, 1931.) order as the series given by Hofmeister' in 1888. As is to be expected, the values of K/ vary from one protein to another in the same salt. b. Effect of Temperature o?i Solubility K/ is, however, independent of temperature and pH. That is, curves for the logarithm of the solubility in the same salt are parallel to each other at various pH values as are also curves at various temperatures. This is apparent in figure 2." Hemoglobin is more soluble at 0° than at 25° C. at high salt concentrations, which is the reverse of the effect of temperature on solubility in low concentra- tions of salt. This phenomenon is also exhibited by other proteins, including the least soluble of the pseudoglobulins from serum. The data of S0rensen and H0yrup" on the solubility of egg albumin in ammonium sulfate solutions at various temperatures again show K/ to be independent of temperature, and again the solubility is greater at 0° C. than at 20° C, but increases at higher temperature. Thus, for a given protein, variation in K/ describes the effect of the type 2 1 8 Solubility o£ Proteins of electrolyte on the solubility, while variations in /? describe the effect of temperature and pH on the solubility in a given salt solution. /? is the intercept constant and therefore the logarithm of the hypothetical solubility at zero salt concentration. It is "hypothetical" because many proteins are so soluble in water that their solubility cannot be measured, and also because true globulins are insoluble in water and show increased solubility in the presence of low concentrations of salt. Variations in p may be conveniently used to describe variation of solubility with pH. c. Effect of pH on Solubility Proteins become positively or negatively charged on either side of the iso- electric point and these forms are more soluble than the electrically neutral molecule. Solubility is increased by the addition of acid or base and the protein behaves as though it had a divalent charge. This is obviously only an apparent phenomenon, but it makes it possible to describe the solubility of a protein at varying hydrogen ion concentration by the equation S =^^,^ an.^ ^ Kz'K,' (2) ++Protein— Sn Ki'K.J ag.^ 5 is the solubility, 5„ the solubility of the neutral molecule, ajj+ the activity of the hydrogen ions, K/K/ the dissociation constants in acid solution, and K/K^' the dissociation constants in alkaline solution. If these dissociation con- stants are sufficiently far apart in value, solubility in alkaline solution may be described by the equation Protein— ^ S_ ^ K^' K,' (3) ++Protein Sn a and solubility in acid solution by the equation 2 H-' Protein++ S , ag.' (4) ++Protem— Sn Ki'K^' Experimental results at different ionic strengths and different reactions may be successfully analyzed in the region where equation 1 holds (since K/ is independent of pH) by variations in p. This has been done in figure 3, which describes the variation of solubility of both hemoglobin in phosphate buffers^ and egg albumin in ammonium sulfate (S0rensen and H0yrup^), with pH. ^n is the logarithm of the hypothetical solubility of the neutral molecule of the protein at zero salt concentration. Thus, S _ S _ antilog /3 1 ++Protein Sn antilog ;8„ The curves drawn through the points are theoretical according to equation 2. The values for 5„ or Pn are not the values for minimum solubility as actually Arda Alden Green 219 determined, but are corrected for slight amounts of both negatively and positively charged molecules present with the neutral molecule even at the pH of minimum solubility. The solubility of casein at various pH values may also be described by equation 2. This equation has been applied success- fully to the data of Linderstr0m-Lang and Kodama* and of S0rensen and Sladek.^" Other Methods of Precipitation Temperature, pH, isoelectric point, and concentration of solvent must also be considered when other methods of separation or precipitation of proteins 40 _L oEQCi ALBUniN 8.7 55 • HEMOGLOBIN 1220 1335 _L A, 5.75 2.77 4.5 5.0 pH OF EGG ALBUMIN SOLUTIONS 5.5 BO 5.5 7.0 pH OF HEMOGLOBIN 50LUTI0N5 7.5 Fig. 3. The solubility of hemoglobin and of egg albumin in concentrated salt solutions of varying pH. (From Green, A. A.: Jl. Biol. Chem. 93:524, 1931.) are employed. These other methods will not be discussed here in any detail. They include colloidal adsorption; differential denaturation, especially by heat; combination with specific substances; or precipitation by such organic solvents as acetone and methyl or ethyl alcohol. Alcohol at low temperatures has been used most recently by Cohn and his co-workers" for the separation of plasma proteins. Antibodies may be specifically precipitated by combi- nation with the antigen. Heidelberger and KendalP have prepared pure pneumococcus antibody by precipitation with the type-specific carbohydrate and subsequent solution in salt solutions. Northrop" has recently crystallized diphtheria antitoxin by digesting the toxin from the toxin-antitoxin complex with the aid of trypsin and subsequently precipitating fractionally with ammonium sulfate. All of the above-mentioned methods may be used in combination with or alternately with isoelectric or salt precipitation. 2 20 Solubility of Proteins Effect of Other Proteins on Solubility The solubility of a given protein in a mixture of other proteins depends not only upon the factors already considered, but also upon the presence and con- centration of other proteins. These differences in solubility are difficult to explain adequately but are undoubtedly due partially to the effect of the other proteins on the electrostatic characteristics of the medium or possibly to the combination of one protein with another. Whatever the reason, pro- teins are usually more soluble in the presence of other proteins. Thus, pro- teins completely precipitated in a given salt solution in the presence of other proteins and other organic molecules found in an extract may require a much lower concentration of salt for precipitation as purification proceeds. After a protein has been purified, the concentration of salt at which the solubility will reach a given low value will be constant, but the point at which precipitation first begins necessarily depends upon the concentration of the protein. Precipitation starts when the concentration of the protein exceeds the solubility of the protein under the given conditions, that is, when the solution is saturated with respect to the protein. If pure proteins are precipi- tated from a more concentrated protein solution than that found in the original extract or mixture, beginning precipitation at a lower salt concen- tration may be due to increased concentration of the protein itself rather than to the removal of other proteins. One other consideration might be mentioned. Comparable proteins from different species do not necessarily have the same solubility characteristics. For instance, crystalline hemoglobin from human red cells, although having the same molecular weight and the same oxygen-combining capacity as that from the red cells of the horse, is one thousand times more soluble in phosphate buffers of the same ionic strength and pH." Thus, it is usually not possible to employ identical purification procedures in dealing with comparable material from different species. Criteria for Purity Since the protein molecule is so large and since its actual chemical structure is still a mystery, the tests for the identity and purity of a given protein fraction are necessarily indirect. One of the most rigorous tests for homogeneity of a given protein fraction is constant solubility under given conditions inde- pendent of the amount of saturating body. Constant chemical composition upon reprecipitation of proteins containing an inorganic constituent such as iron, copper, sulfur, or phosphorus is also an excellent indication of purity. One boundary in the electrophoresis cell and one molecular weight by diffu- sion and sedimentation measurements, constant mobility, dielectric constant, amphoteric properties, and viscosity all give additional information. Enzymes, hormones, antibodies, and other proteins having physiological functions which exhibit constant activity per milligram of protein upon repeated refrac- Arda Alden Green 221 tionation may be assumed to be relatively pure. Ultimately the most important single factor in the isolation of a given protein is an adequate method of char- acterizing and identifying that protein. The Plasma Proteins The isolation of individual proteins from plasma is one of the most pressing of present-day problems in biochemistry. Proteins with peculiar physiological roles such as fibrinogen, prothrombin, the recently isolated protein constitu- ents of complement,^" the enzymes choline esterase and phosphatase, and the antibodies to various bacteria are more easily separated, since one has the biochemical yardstick as well as the ordinary means of chemical or physico- chemical characterization. It is no longer adequate to make a separation into "globulin" and "albumin" fractions. The application of the new cataphoretic methods of Tiselius" has divided the globulin fraction of serum into at least three components a, ^, and y, and the albumin fraction into two." Extensive studies have been carried out correlating ammonium sulfate and alcohol precipitation methods with elec- trophoretic mobility and with molecular weight as determined by ultracentri- fuge and diffusion methods of both the albumin and globulin fractions."" Cohn's recent review'" divides the plasma proteins into groups based on their electrophoretic mobility. It is becoming increasingly evident that each new method of identifying or classifying serum proteins increases the ntunber known. The simple pro- cedure of separating the water-soltible from the water-insoluble fractions of the globulins is an example of such a procedure. Three reproducible iso- electrically precipitable fractions of water-insoluble globulins have been separated from normal horse serum, and their amphoteric properties and viscosities studied.-"'^ They have been called Pj, Pu and Pm. Pn and Pi„ are the more insoluble and have isoelectric points at 6.2 and 5.0 respectively. Reiner and Reiner" had previously found two euglobulin fractions in normal and in antipneumococcus horse serum which correspond to Pu and Pm- Hewitt^ has described two euglobulin fractions obtained from sera from a number of species which he calls euglobulin I and II with isoelectric points at pH 7 and pH 6. Svensson"* has prepared these various fractions and com- pared their electrophoretic mobility with the following results: "The frac- tions euglobulin I and II, described by Hewitt, were found to behave dis- tinctly different in electrophoresis. The former contained principally f3- and y-globulins; the latter was a rather homogeneous a-globulin preparation. . , . The fraction Pi of Green was shown to be identical with euglobulin II of Hewitt and with a-euglobulin of Tiselius. The results indicated that Green's fractions P„ and Pm correspond to y- and /3-euglobulins respectively" (p. 824). The blood proteins, apart from their various specific functions and their general function of maintaining the osmotic pressure, apparently combine with smaller organic molecules including phospholipids and carbohydrates. 222 Solubility o£ Proteins Chick,"^ S0rensen," and others have studied the phosphorus content of serum proteins as a measure of the phosphoHpids present. The phospholipid is largely combined with the water-insoluble proteins. The phosphorus content may be higher than 0.3 per cent, which corresponds to 7.5 per cent of phospho- lipid. MacheboeuP has described a protein containing a much higher percent- age of phospholipid as well as cholesterol. Cholesterol has also been reported in both the euglobulin and albumin fractions."' ^■^'' KendalP reports the pres- ence of a fatty acid associated with crystalline human albumin "that cannot be removed without first destroying the albumin" (p. 109). TABLE 1 Galactose-Mannose Content of Horse Serum Globulins Rimington^" (i9,'!i) S(zirensen and Haugaard^' (1933) Tiselius'* (i937) Hewitt23 (1938) Per cent of carbohydrate 2.0 1.8 0.7 2. 2 0.4 1-4 2.4 5.6 Fraction analyzed Total globulins Total globulins Gamma globulin Beta globulin Alpha globulin Pseudoglobulin A Main pseudoglobulin Globoglycoid The carbohydrate bound to serum proteins is reported to be probably galactose-mannose-acetylglucosamine.'^'^" The only protein free from bound carbohydrate is one of the crystalline albumins; the other crystalline albumin contains about 5 per cent of carbohydrate; and Rimington^^ describes the noncoagulable protein from serum, seromucoid, as containing 10.7 per cent of carbohydrate. Various authors have investigated the carbohydrate content of globulins and their results on horse serum are given in table 1. The most extensive survey of the carbohydrate content of serum proteins has been carried out by Hewitt,^' ^^ who divides the globulins from the serum of a number of species into his two euglobulin fractions and at least three pseudoglobulins, the most insoluble. A, the main fraction, and globoglycoid which contains up to 5 per cent of carbohydrate. We^^ have obtained similar results with a more extensive fractionation. The pseudoglobulin may be di- vided into more fractions and, as Hewitt suggests, it is possible to obtain pseudoglobulin with a somewhat lower carbohydrate content. However, none have been found to be carbohydrate free. It is evident that proteins really combine with these smaller organic mole- cules as Hewitt^ says: "That the polysaccharide present in protein forms an integral part of the protein molecule and is not merely a loosely bound con- Arda Alden Green 223 taminant is shown by the difficulty of removing this bound carbohydrate. Prolonged fractionation with salts, isoelectric precipitation, dialysis, heat coagulation and even specific flocculation between antigen and antibody all failed to remove the carbohydrate present in serum proteins" (p. 1502). If one studies combination with any one substance, the proteins may be relatively easily separated into groups having a constant percentage of that constituent. If more than one compound is considered the problem is much more difficult. For instance, the number of euglobulins of serum are multi- plied if one considers both phospholipid and carbohydrate content. Un- doubtedly, there are also other substances combining with proteins. Proteins combine not only with other substances but with each other. The problem of the identification of the euglobvdins is further complicated by the ability of water-soluble globulins to unite and form water-insoluble globulins behaving like Pj. This was first recognized by Kendall,^* working on human serum. He noted a precipitate formed on the addition of the less soluble pseudoglobulin to the more soluble pseudoglobulin. He showed quantitative relationships depending upon pH, analyzing the protein content by im- munological means. Hewitt" has since shown that his pseudoglobulin A from human serum will form a precipitate with the globoglycoid separated from the serum of different species. It is evident that pH is especially important in trying to separate these two proteins. Ever since S0rensen's* classical studies on the solubility of crystalline serum albumin it has been known that there must be more than one serum albumin, since solubility was not independent of the amount of solid crystals present. Two albumins have now been crystallized from horse serum having different electric mobility and different carbohydrate content. One is carbohydrate free and one contains 5.5 per cent of carbohydrate.^" ^"^ The carbohydrate-free protein may also be crystallized from concentrated salt-free solutions as the sulfate. "° Although successful studies on the solubility of fibrinogen had been carried out previously/^ the carbohydrate-free albumin is the only protein so far isolated from serum exhibiting constant solubility independent of the amount of saturating body. Carbohydrate-free crystalline human serum albu- min shows the same evidence of purity.^ Other proteins so far found in the albumin fraction are enzymes, including choline esterase and phosphatase, a crystalline copper containing protein, hemocuprein,^' seromucoid,"' and a certain amount of euglobulin. The solubility curves of these serum proteins necessarily overlap tremen- dously. Advantage must be taken of all possible variations in (i and K^. That is, differences in variation of solubility with temperature, with pH, with kind and concentration of electrolyte, and with type of solvent must all be utilized. Separation can be effected only after repeated fractionation, the precipitating agent being added in small increments under controlled conditions, including control of the protein concentration, and careful analysis of each fraction be- fore pooling with other similar fractions. 2 24 Solubility of Proteins After one has succeeded in obtaining homogeneous fractions of constant chemical composition and constant physicochemical characteristics it is dis- heartening to remember that not only will similar fractions from another species be different, but probably fractions obtained in the same manner from serum of another animal of the same species or even from serum of the same animal under different conditions will vary considerably. The recent work of Schoenheimer and co-workers^" on the heavy nitrogen containing amino-acid content of antibody-protein upon injection of the heavy nitrogen after the antibodies have been completely formed demon- strates the constant fluidity of the protein molecule even in regard to its amino-acid composition. In fact, it is surprising that the composite picture of serum proteins has such fundamentally stable and definite outlines that reproducible fractions can be obtained. REFERENCES 1 . Cohn, E. J.: Physiol. Revs. 5:349, 1925. 2.S0rensen. S. P. L., and H0yrup, M.: Cpts. rd. Laborat. Cailsberg 12:213, 1915-1917. 3. Chick, H., and Martin, C. J.: Biochem. Jl. 7:380, 1913. 4. S0rensen, S. P. L.: Jl. Amer. Chem. See. 47:457, 1925; Cpts. id. Laborat. Carlsberg 18(5), 1930; Kolloid Ztschr. 53:102, 170, 306, 1930. 5. Green, A. A.: Jl. Biol. Chem. 95:47, 1932. 6. Green, A. A.: Jl. Biol. Chem. 93:495, 1931- 7. Hofmeister, T.: Naunyn-Schmiedebergs Arch. f. exper. Pathol. 11. Pharmakol. 24:247, 1887-1888. 8. Green, A. A.: Jl. Biol. Chem. 93:517, 1931. 9. Linderstr0m-Lang, K., and Kodama, S.: Cpts. rd. Laborat. Carlsberg 16(1), 1925. 10. S0rensen, S. P. L., and Sladek, L: Cpts. rd. Laborat. Carlsberg 17(14), 1929. 11. Cohn, E. J.; Luetscher, J. A., Jr.; Oncley, J. L.; Armstrong, S. H., Jr.. and Davis, B. B.: Jl. Amer. Chem. Soc. 62:3396, 1941. 12. Heidelberger, M., and Kendall, F. E.: Jl. Exper. Med. 64:161, 1936. 13. Northrop, J. H.: Jl. Genrl. Physiol. 25:465, 1942. 14. Green, A. A.; Cohn, E. J., and Blanchard, M. H.: Jl. Biol. Chem. 109:631, 1935. 15. Pillemer, L.; Ecker, E. E.; Oncley, J. L., and Cohn, E. J.: Jl. Exper. Med. 74:297, 1941. 16. Tiselius, .\.: Biochem. Jl. 31:313, 1464, 1937; Trns. Faraday Soc. 33:524, 1937. 17. Luetscher, J. A., Jr.: Jl. Amer. Chem. Soc. 61:2888, 1939. 18. Cohn, E. J.; McMeekin, T. L.; Oncley, J. L.; Newell, J. M., and Hughes, W. L.: Jl. Amer. Chem. Soc. 62:3386, 1941. ig. Cohn, E. J.: Chem. Revs. 28:395, 1941. 20. Green, A. A.: Jl. Amer. Chem. Soc. 60: 1 108, 1938. 21. Fahey, K. R., and Green, A. A.: Jl. Amer. Chem. Soc. 60:3039, 1938. 22. Reiner, H. K., and Reiner, L.: Jl. Biol. Chem. 95:345, 1932. 23. Hewitt, L. F.: Biochem. Jl. 32:1540, 1938. 24. Svensson, H.: Jl. Biol. Chem. 139:805, 1941. 25. Chick, H.: Biochem. Jl. 8:404, 1914. 26. Macheboeuf, M.: Bull. Soc. de chim. biol. 1 1:268, 485, 1929. 27. Gardiner, J. A., and Gainsborough, H.: Biochem. Jl. 21:141, 1927. 28. Handovsky, H.; Lohman, K., and Bosse, P.: Pfliigers Arch. f. d. gsmt. Physiol. 210:63, •925- 29. Kleczkowski, .\.: Biochem. Ztschr. 299:31 1, 1938. Arda Alden Green 225 30. Kendall, F. E.: Jl. Biol. Chem. 138:97, 19.) i. 31. Hewitt, L. F.: Biochem. Jl. 33:1496, 1939. 32. Riinington, C, and Van den Ende, M.: Biochem. Jl. 34:941, 1940. 33. Green, A. A.; Fahey, K. R., and Green, S. V.: Jl. Biol. Chem. i4o:xlvii, 1941. 34. Kendall, F. E.: Cold Spring Harbor Sympsa. Quantit. Biol. 6:376, 1938. 35. McMeekin. T. L.: Jl. Amer. Chem. Soc. 61:2884, '939- 36. McMeekin, T. L.: Jl. Amer. Chem. Soc. 62:3393, 1940. 37. Florkin, M.: Jl. Biol. Chem. 87:629, 1930. 38. Mann, T., and Keilin, D.: Proc. Roy. Soc. London, s.B. 126:303, 1938. 39. Schoenheimer, R.; Heidelberger, M.; Riltenberg, D., and Ratner. S.: Jl. Biol. Chem. i4o:cxii, 1941. 40. Rimington, C: Biochem. Jl. 25:1062, 1931. 41. S0rensen, M., and Haugaard, G.: Cpts. rd. Laborat. Carlsberg 19(12), 1933. GONADOTROPHIC STIMULATION OF THE OVARIES OF THE ADULT RHESUS MONKEY By CARL G. HARTMAN X From the DEPARTMENT OF EMBRYOLOGY CARNEGIE INSTITUTION OF WASHINGTON BALTIMORE, MARYLAND CxONADOTROPHIC STIMULATION OF THE OVARIES OF THE ADULT RHESUS MONKEY SINCE SMITH in 1926 and Aschheim and Zondek about the same time dis- covered the gonadotrophic function of the mammalian anterior pituitary gland (a logical sequence to Smith's epoch-making work, done in California, on the primacy of the anterior pituitary in the tadpole), a deluge of literature on the pituitary-gonad relation has appeared. Some idea of the researches in this field may be gained by a perusal of volume 1 1 of Memoirs of the Univer- sity of California (1932) on "The Growth and Gonad-stimulating Hormones of the Anterior Lobe of the Hypophysis" by H. M. Evans and co-workers. The progiess since that date may further be gauged by comparing the 1939 and the 1932 editions of E. Allen's compilation, "Sex and Internal Secretions." When one surveys the literature on the gonadotrophic action of anterior pituitary extracts one is struck with the paucity of references to ovulation, barring, of course, studies on the Friedman test as performed on that aberrant mammal, the rabbit, which has more than once in the last hundred years brought temporary confusion into biological studies. A review cannot be made here even of the few studies devoted to ovulation in mammals,' but some of the more important references are given in the literature list below. Ovulation is still too uncertain an end point to be serviceable in assaying an unknown extract for its gonadotrophic value. This is due to individual as well as strain differences of susceptibility within a species and to very wide differences between species. This is especially marked with respect to human urine of pregnancy (PU), less so with pregnant mare's serum (PMS), and least of all with follicle-stimidating hormone (FSH) made from the gland itself. Thus far no one has been consistently successful in producing ovulation in the adult rhesus monkey. Hisaw and collaborators' brought about OAula- tion in young animals but the gonadotropes suitable for ovulation in adult females have, so far as the writer is aware, not yet been found. Almost without exception a diffuse overstimulation of the ovaries has been produced with gonadotropes, but never an ovulation, except for the several reported by the writer in 1938. After the preliminary demonstration of induced ovulation in prepubertal rats by means of anterior pituitary transplants'* search was made for the active principle or principles responsible for the action of the pituitary on the gonads, notably in the laboratories of Hisaw and Fevold, of Evans, and of Van Dyke. This interesting story has often been told and will not be re- peated here; suffice it to say that the dualistic theory, despite its difficulties, is still in the ascendency. By fractional precipitation it has been possible to re- move from crude anterior pituitary extracts, in a highly purified state, one [229] 230 Gonadotrophic Effects in Rhesus fraction which principally causes growth of the Graafian follicle, another which it is necessary to add to effect its rupture. Since 1934 the writer has been interested in the clinical problem of increas- ing the incidence of ovulation in the menstrual cycles of certain hypotypical females of the Carnegie rhesus colony, for embryos were desired,^ and ovula- tion is a prerequisite for conception. The results of 104 experiments were pub- lished in 1938,^ but they do not inspire enthusiasm. In 104 experiments only seven ovulations were observed which could reasonably be attributed to the treatment. It is now possible to report 41 further experiments performed on 33 females in the last three years. The results were critically controlled as be- fore by a prior study of the subjects, by palpation, by laparotomy, and by histological examination of biopsied or autopsied material. The following variations in technique were instituted: both lower and higher dosages were employed; injections were begun a week before the next expected menstrual period, so as to allow an additional week before the middle of the following menstrual interval for the slow building up of the ovaries; intravenous per- fusion over a period of eight hours was tried twice; estrogens were injected concomitantly with the gonadotropes; besides Gonadogen Upjohn (PMS), good anterior pituitary extracts (AP) were made available. One batch of FSH was prepared by Dr. H. Jensen of E. R. Squibb and Sons and used alone (with good results), or with Follutein Squibb (PU-APL) in proportion of 1:2 or 1:10. In the last proportion the combination acted like APL itself, which Engle" ' long ago showed to be inert in the monkey. A second potent AP extract rich in FSH was furnished by the Abbott Com- pany through Dr. Hazel. The extract had been exposed briefly to tryptic diges- tion after the method of McShan and Meyer, presumably to increase the FSH:LH ratio by differential digestion of the LH. It should be emphasized that in all cases subjects were selected which were (and because they were) not ovulating at the time of the experiments, hence were presumably unreactive to their own intrinsic hormones. Results The results of 38 experiments done on 30 animals may be tabulated as follows: Extract Ovulations Overstimulation Without effect Gonadogen 3 7 5 Gonadogen, 1,000 U.U.* o 3 o FSH — Jensen 1 2 o FSH — Jensen + Follutein i:2.... o i o FSH — Jensen + Follutein 1 : 10. . . 009 FSH— Abbott 121 Gonadotrope + Estrogen o i 6 6 16 21 * Upjohn units, or Cartland-Nelson units (lo U.U. = loo Cole-Saunders units): the minimum total dose which, administered to 21- to 23-day-old rats weighing 3S~4S gni., in ^ equal subcu- taneous injections at daily intervals, will produce at autops/, 96 hours after the first injection, a mean ovarian weight of 65 mg., which is 4-5 times that of the controls (Amer. Jl. Obstet. & Gynecol. 38:1024, 1939). Carl G. Hartman 231 Ovulatiojis. Of the six ovulations observed and credited to the action of the hormones, one is somewhat doubtful. One case deserves special notice: No. 584 had not ovulated while in our possession up to Jan. 1940. An exploratory laparo- tomy was performed on Jan. 25, day 12 of the cycle, ^vhen a 4x3-0101. follicle was found in the right ovary; 10 II. U. of Gonadogen Avere at once injected intravenously. The next day the female was mated. On Feb. 7 a 12-day ovum (C637) was recovered by hysterotomy. This case raises some interesting points. It is probable that in any species ovulation may be precipitated by a correct intravenous dose of gonadotropes provided a large follicle is present in the ovary. Hence it is likely that in clini- cal experience some of the successes (induced ovulations) may have been merely cases of slightly hastening the ovulation. Since women cannot be laparotomized (as was monkey No. 584 before a course of treatment merely for an inspection of the ovaries), the proper control is lacking in such gyneco- logical experiments. The case further illustrates the spontaneous resumption of ovarian function after a period of low activity, a phenomenon which has been seen so often in the Carnegie colony that we have become extremely cautious in evaluating our results. It is significant that but one ovulation was effected in March, the beginning of the nonbreeding (refractory) season; the others were referable to the fall and winter, the "best" season for the rhestis monkey. In three cases 1,000 U.U. of Gonadogen were injected subcutaneously on June 30, 1941; in no case did ovulation result, but only overstimulation of the ovaries. Diffuse Overstimulation of the Ovaries. In the event of failure to effect ovula- tion, if the introduced gonadotropes have any effect at all, it is to awaken an overabundance of follicles to growth, which results in excessively large glands of "cystic" appearance. In these ovaries the liquor folliculi is very dense and, despite the large total quantity, there is evidence that generally the estrin output is very low. The Graafian follicles are like overstimulated and exhausted thyroid alveoli which are rich in colloid but poor in thyroxin. The color of the sex skin often be- comes brighter and the vaginal desquamation increases as a result of the injections, but far below that expected. Removal of one ovary does not hasten the onset of the sticceeding menstrual flow appreciably and removal of both does so but slightly. One of the evident effects of the treatment with gonadotropes on the ovary is the congestion, which is in many cases visible externally as great, raw-flesh colored areas overlying dilated follicles and, in one ovulatory case, overlying the normal corpus luteum. It was at first believed that these regions were denuded of ovarian cortex; but sections showed them to be due to dilated blood vessels in the somewhat thinner cortex. Internally, this congestion is everywhere apparent in the thecal vessels of follicles Avhich in the living ani- 232 Gonadotrophic Effects in Rhesus mal, as viewed directly during an exploratory laparotomy, often appear as red rings about the numerous growing iollicles. Such evidences of overstimulation are occasionally met with in untreated animals, in which the ovary grows excessively by virtue of the abnormally large number of follicles, the surface phenomena of congestion just described, and blood follicles. This condition may persist over a number of menstrual periods. The excessive congestion inevitably leads to degeneration of the grandulosa and soon to regression of the follicles which is extremely rapid, as often de- termined by palpation, twice corroborated by successive exploratory laparo- tomies and frequently by a single laparotomy after the treatment. The rapid regression of the ovaries was reflected in the premature onset of menstruation, which frequently occurred as early as 4 to 6 days after the regression became evident, shortening the cycle to 19, 20, or 21 days. Occasionally the expected bleeding did not occur but, instead, a long period of amenorrhea set in. This, again, was neatly related to the time of year, for amenorrheic periods were likely to follow treatment in March and rarely occurred from October to January, when recovery was generally prompt; the treated female resumed normal menstrual cycles and thereafter ovidated spontaneously. The nonpredictability of the adult rhesus monkey may, however, be illus- trated by the following bizarre case. After a series of anovulatory cycles female No. 678 ^vas (in March) treated with FSH-Jensen -I- Follutein 1:10, with little or no effect. No bleeding took place until June 10. On June 23, which is well within the anestrum or non-breeding season of the rhesus monkey, she con- ceived! She carried her baby to term and nursed it normally. Injections without Effect. As already pointed out, refractoriness of the adult rhesus monkey to gonadotropes becomes greater as the nonbreeding season characteristic of the species approaches, namely the spring of the year. Normal females which suddenly stop ovulating may become absolutely unresponsive to huge quantities of the hormone. Animals whose ovaries are unresponsive to their intrinsic hormones in the fall and winter are usually not to be stimu- lated by exogenous gonadotropes. But even in these one meets with the great- est surprises. In other words, the writer has not been able effectively to predict, from a prior study of the animal, what the effect of treatment might be, except, of course, in the case of obviously sick individuals. Antihormones play no part when such small dosages are administered over so short a period of time, for it has happened that the second treatment with the same material has met with greater response than the first.^ Kind of Hormone. Gonadogen again proved itself an active stimulator of the monkey ovary and seems on the whole a very good gonadotrope. FSH- Jensen and FSH-Abbott made a very good showing in the small series of experiments here reported, when used without the admixture of APL (Follu- tein). When Follutein was added to Jensen's FSH in the proportion of 2 of Follutein to 1 of FSH, the results seem less favorable, but yet stimulation re- Carl G. Haitman 233 suited; when the proportion was raised to io:i the combination became as inert as APL by itself. The reverse proportion, lo units of FSH to i of Piolan (APL), as used effectively by Evans and Simpson, might have been ideal.* Combined Action of Estrin and Gonadotropes. The small series of seven ex- periments under this heading are, of course, not fully conclusive, but, since six out of seven trials gave completely negative results, nothing can be said in favor of the method used. Certainly congestion of the genital tract was accom- plished by the estrogens in these experiments, as shown by the enlargement of the uterus in most of the subjects; hence the idea which originally prompted the experiment did not work out. The single animal (No. 217) which responded excessively to the combined hormones differed in no essential from two nega- tive cases (Nos. 215 and 220) similarly treated at the same time, except that the estrogens were left out. The remarkable discovery of Pencharz," however, confirmed by Simpson, Evans and co-workers," of the synergisms of estrogens and gonadotropes in the hypophysectomized rat, would seem to demand further experimentation with this method in the monkey. In the light of these findings we come back to the original method of Hisaw, Fevold, and collaborators, namely using FSH to build up the follicles, then following with an "ovulatory dose" of FSH to which LH has been added. For women and the monkey female it seems that we must argue against the use of APL (FoUutein and Antuitrin S and related products) except perhaps in minimal qtiantity in connection with FSH which had been previously tested in hypophysectomized rats. Thanks are due to the Squibb Institute for Medical Research and the research labora- tories of the Abbott Company and the Upjohn Company for generous cooperation in these experiments. REFERENCES 1. Hartman, C. G.: in Sex and Internal Secretions, ed. by E. Allen (2d ed.; Baltimore: 1939), 630. 2. Fevold, H. L.: in Sc\ and Internal Secretions, ed. by E. Allen (2d ed.: Baltimore: 1939). 966. 3. Smith, P. E., and Engle, E. T.: Amer. Jl. Anat. 40: 159, 1927. 4. Heuser. C. H., and Streeter, G. L.: Contribs. Embryol., Carnegie Instn. Washington 29:15, 1941- 5. Hartman, C. G.: Bull. Johns Hopkins Hosp. 63:351, 1938. 6. Engle, E. T.: Amer. Jl. Physiol. 108:528, 1934. 7. Engle, E. T.: Amer. Jl. Physiol. 106:145, 1933. 8. Evans, H. M., and Simpson, M. E.: Endocrinologv 27:305, 1940. 9. Pencharz, R. I.: Science (n.s.) 91:554, 1940. 10. Simpson. M. E.; Evans, H. M.; Fraenkel-Conrat, H. L., and Li, C. H.: Endocrinology 28:37, 1941. Relevant Literature Allen, E., editor: Sex and Internal Secretions (2d ed.; Baltimore: 1939). Burdick, H. O., and Whitney, R.: Amer. Jl. Physiol. 132:405, 1941. Cole, H. H.: Amer. Jl. Physiol. 119:704, 1937. Cole, H. H., and Miller, R. P.: Amer. Jl. Physiol. 104: 165, 1933. Engle, E. T.: in Sex and Internal Secretions, ed. by E. Allen (2d ed.; Baltimore: 1939), 1003. 234 Gonadotrophic Effects in Rhesus Erving, H. W.; Scnis, C, and Rock, J.: Amcr. Jl. Obstet. &: Gynecol. 40:695, 1940. Frank, R. T., anci Berman, R. L.: Endocrinology 28:211, 1941. Hisaw, F. L.; Creep, R. O., and Fevold, H. L.: Anat. Recrd. 6i(supp.)3:24, 1935. Huber, C. P., and Davis, M. E.: Surg., Gynecol. Sc Obstet. 70:996, 1940. I-cathem, J. H.: Endocrinology 22:559, 1938. Leathern, J. H., and Starky, W. F.: Proc. Pennsylvania Acad. Sci. 14:109, 1940. Nelson, W. O., and Overholser, M. D.: Jl. Pharmacol. & Exper. Therapeut. 54:378, 1935. Pincus, G.: Anat. Recrd. 77:1, 1940. Smith, P. E.: Jl. Amer. Med. Assn. 88: 158, 1927. van Wagenen, G., and Morse, A. H.: Lancet 234:1220, 1938. Windle, W. F.: Endocrinology 25:365, 1939. THE PATHOLOGIC, CLINICAL AND BIOCHEMICAL CORRELATION OF TUMORS OF THE TESTIS By FRANK HINMAN, A.B., M.D., F.A.C.S. X From the DEPARTMENT OF SURGERY, DIVISION OF UROLOGY UNIVERSITY OF CALIFORNIA MEDICAL SCHOOL SAN FR.\NCISCO, CALIFORNIA THE PATHOLOGIC, CLINICAL AND BIOCHEMICAL CORRELATION OF TUMORS OF THE TESTIS Introduction TUMORS of the testis always have presented a confusing picture. Their patho- genesis remains uncertain and they have never been classified satisfactorily. When the contemporary pathologic, clinical, and biochemical contributions, however, are brought into relation with each other, a clearer picture is pre- sented than ever before. As elsewhere a neoplasm may arise in the testis from any tissue cell present there and may be benign or malignant. For practical purposes, however, they are all malignant, so exceptional are the benign tumors. In earlier pathologic studies homologous or monocellular growths were distingiushed from heterol- ogous or multicellidar growths and in each group certain types were defined. At first, the tumor of the one-cell pattern was identified commonly as a sar- coma. Now we know that a pure mesoblastic growth of the testis is very rare. These same monocellular tumors were next designated as seminomata or spermatocytomata, unrelated to sarcoma, and supposedly originating from sperm cells. More recently they have been called either embryonal carcinoma, with or without lymphoid stroma, or large clear-cell carcinoma, and the origin of any of them from adult sperm cells is questioned. The tumors of variegated- cell pattern, the mixed or heterologous growths called teratomata, were grouped as benign or malignant and the latter were subdivided according to the type of malignant cell. Such types, for example, as adenocarcinoma, spheroidal-cell scirrhous carcinoina, etc., and chorionepithelioma were rec- ognized. This dual classification with its subdivisions proved most unsatisfac- tory. Now and then a tumor appeared which failed to fit into any part of the classification. Years ago Ewing suggested that all testicular tumors, with the obvious few exceptions, probably arose from a totipotent cell. Recent clinical observations confirm this view^ Perhaps tumors of the testis may better be regarded by the clinician not only as malignant but as being one tvpe of growth, the varieties arising from the different stages or phases of transition at which they are observed. In teratoma, groups of cells are found which are identical with the common unicellular types, and frequently careful exami- nation of serial sections of the supposedly pure-cell type uncovers other \ ari- eties of cell, thus proving its heterogenicity. Distinction between the one-cell and the mixed-cell patterns serves no practical purpose. Rather, confusion arises in attempting such distinction. Of much more clinical value is the microscopic recognition of primitive or adult characteristics. The primitive cell is embryonal in appearance, the adult cell seems fully differentiated. Even C 237 1 2q8 Tumors o£ the Testis this distinction is not always clear-cut. There are transitional stages. The tumor may seem to be solidly composed of fully differentiated adult cells until diligent search through serial sections discloses undifferentiated pri- mordial cells or scattered gioups of such cells which may be in different stages of transition. The most primitive form of cell is the one by which the tumor should be classified. Chorion-like cells are the most premature. Whether this distinguishing cell occurs in a monocellular or in a mixed type of tumor is unimportant. In either case the growth is a chorionepithelioma. The recog- nition of the embryonal or mature character of a tumor with transitional phases in either direction has considerable clinical value as can be demon- strated by correlation of the pathologic, clinical, and biochemical facts. Pathology Ewing grouped all tumors of the testis in four ways: 1. Adult embryoma or teratoma (benign) 2. Embryonal teratoid or mixed tumor (malignant) 3. Embryonal malignant tumor (embryonal carcinoma) 4. Miscellaneous (rare) a. Adenoma h. Fibroma c. Myoma d. Sarcoma e. Interstitial-cell tumor /. Lymphosarcoma g. Adult multicystic adenocarcinoma (arising from adult cells of semi- niferous tubules) The tumors which can be classified under groups 1 and 4 are very rare, so rare indeed that, clinically, only groups 2 and 3 need be considered in the present correlation of facts. Disregard of the distinction between monocellular and mixed-cell tumors unites these two groups so that, except for the fore- going very rare neoplasms mentioned in groups 1 and 4, there is only one testicular tumor. This tumor probably arises from a totipotent cell and may undergo a gieat variety of transitions, the nature of which can be inferred by the microscopic picture presented at the particular phase present when the tumor is examined. If the growth is allowed to develop further, however, subsequent examination may disclose an entirely different picture. For ex- ample, the primary growth found in the testis after castration may show a pure embryonal carcinoma but the metastases examined months later may disclose chorionepithelioma. Suspecting the importance of cellular maturity as the basis of classification of testicular tumors. Dr. Jesse L. Carr, who has reviewed all our pathological material, attempted such a differentiation several years ago. Careful correla- tion of pathological with clinical and biochemical facts has evolved the fol- lowing division, the principal distinction of which is recognition of cellular Frank Hinman 239 maturity. Any complete classification, of course, must include the rare tumors of Ewing's groups 1 and 4. All tumors in his groups 2 and 3, however, can be listed as follows under parenchymatous tumors arising in a teratoma. a. Embryonal carcinoma without lymphoid stroma b. Embryonal carcinoma with lymphoid stroma c. Large clear-cell carcinoma d. Embryonal adenocarcinoma e. Chorionepithelioma The most common and probably least malignant tumor of the testis is the embryonal carcinoma, of which there are two types. Usually this is a mono- cellular tumor of rapid growth (the seminoma of Chevassu). a. The embryonal carcinoma without lymphoid stroma is a tumor with well-differentiated cells resembling adult seminal epithelium. In the micro- scope large, pale, round or polyhedral cells of uniform size are seen lying in a light fibrous stroma containing few lymphoid cells. Occasionally an alveolar arrangement is observed. Many mitotic figures are present. This tumor is relatively rare. b. The embryonal carcinoma with lymphoid stroma is the most common tumor of the testis. It also is made up of well-differentiated cells. By the micro- scope small dark-staining cells, usually in sheets, are seen lying in an abundant lymphoid stroma composed of immature fibroblasts. c. The large clear-cell carcinoma is made up of less well-differentiated cells than those found in the foregoing tumors. It is probably more malignant. Microscopically, closely packed large cells with relatively large pale nuclei containing dark nucleoli, are grouped in a light fibrous stroma. They resemble somewhat the Langhans type of cell. This tumor is uncommon. d. The embryonal adenocarcinoma occurs in two varieties which are not easily distinguished: 1. The adenocarcinoma, made up of primitive cells, is relatively rare. The characterizing embryonal cell is large, producing a moderate amount of cytoplasm and having a finely divided chromatin. Clinical follow-up of pa- tients with this type of tumor, however, has shown a high degree of malig- nancy and in two cases particidarly Carr recognized definite transitional stages between adenocarcinoma and chorionepithelioma. 2. The more differentiated type of embryonal adenocarcinoma occurs next in frequency to the embryonal carcinoma with lymphoid stroma. The mic- roscopic appearance is similar to that seen in adenocarcinoma of other epi- thelial structures. e. Chorionepithelioma is the most primitive and the most malignant of the testicular tumors. Microscopically it is made up of large undifferentiated types of cell and the characteristic multinuclear syncytial cells are found. Judging from the literature, this tumor is rare, but of the last forty-four cases of tumor of the testis seen here six were chorionepitheliomata. In two of these the pri- mary tumor showed no chorionic elements, but the clinical course and the rise 240 Tumors o£ the Testis in gonadotropic hormone output subsequently suggested this possibility, and this was proved at autopsy in both patients. In a third patient no nodule could be palpated in the testicle but a large retroperitoneal mass was found, as well as gynecomastia, and the urine contained 200,000 R.U. of hormone. Biopsy demonstrated chorioncpithelioma. Biochemistry Careful hormonal studies were made on all our patients both before and after operation. We are indebted to Dr. Herbert M. Evans, who has made all recent assays of the urine, for his interest in this subject and his help in correlating the clinical with the biochemical studies. Briefly, the routine method of testing the urine is to get a collection (for 24 hours or longer) from which the gonadotrophin is concentrated by the Levin-Tyndale method of precipitation with tannic acid. Normal immature female rats, 25 or 26 days old, are used and two groups of three animals each are utilized for the test. Over a period of four days aliquot portions of powder derived from a known amount of urine are injected into each animal and autopsy is performed on the fifth day. A positive test is one which shows follicular or uterine stimula- tion. Each animal of the first group is injected with a concentrated powder equivalent to 100 cc. of native urine. Rats of the second group are injected with the amount of powder derived from 4 cc. of urine. If these rats show no response there are fewer than 10 R.U. per liter of urine. If the animals com- prising the first group show a positive reaction, while those of the second group are not stimulated, there are between 10 and 250 R.U. per liter of urine. If both groups of rats react positively there are over 250 R.U. per liter, and fur- ther tests are performed with varying dilutions of native urine in order to detei'mine accurately the number of units. By this method normal men are found to excrete fewer than 20 R.U. and usually fewer than 10 R.U. per liter of urine. Table 1 gives a summary of the results obtained by the use of this test on 44 recent patients with testicular tumor. Clinically the test has proved of no diagnostic value with this exception: when it shows 10,000 R.U. or more of gonadotrophin per liter of urine, the diagnosis of chorioncpithelioma is justified. So uniformly is the high output related to this type of tumor that in some localities, particularly Copenhagen, it is accepted as a fact that chorion- epithelioma is the only type of tumor which causes the excretion of this hor- mone and that, whenever it is present in abnormal amounts, search in the growth or the metastases will disclose chorionic elements. In those patients in whom the test was positive before castration, subsequent testing of the urine has considerable value in prognosis. When the gonadotrophic output is correlated with both the pathological type of tumor and the presence or absence of mestastases a fairly accurate prognosis is possible. pa < a o S D H a: o H O S z o a, M o o £> >o r4 o ^s Z ^ '"' <^ o a ga .5 S 09 O «-»-. ro VD •t X g o g o o o' o ij _c *c 3 1 "o i_ 4> 8 o X 4J d C. ■■ _c '^ o ^ u o -o rt C o X o 0 Ml „' 0 to "c 3 *j 8 X O u-i i ti bj D 1 ij u z 2 >< X X -a 'o j= -13 S '5 >^ o- 4-1 S 3 >-, O ~^ _c JS ■^ c 'i '? £ 1_ c 0 « rt rt c £ 3 ° 1 0 c ns -a a C3j £ U c u. u ~ P lu 'z: c > .5 D - 2 o o 5 > c 1 L. '£ o „ 4- 1- rs — C 5 rt C O c g p: Q -T3 — ^ '5 "rt n o C c 4^ o £ c 4- P u c ?. E '^ >- C 0; c H Si w W < u 0. « H CL, ^ - c« c a- -c 24-2 Tumors o£ the Testis Clinical Although found at all ages, tumors of the testis occur most frequently in early adult life, mostly between the ages of 20 and 40 years. As might be ex- pected, most of the tumors occurring in infants and boys are of the mixed- cell type (teratomata). As a rule all tumors run a rapid course and death is frequent within a year from the time of discovery. Occasionaly a malignant timior has been present for two or three years before the patient has sought medical advice, but the accuracy of the patient's observation in respect to onset may well be questioned as there may have been another condition to start with and the tumor have appeared secondarily. On rare occasions the primary growth in the testis could not be discovered on physical examination before metastases appeared. Some of these patients came to autopsy and the true nature of the metastases were then discovered. In a few instances, in our own experience, biopsy of metastases disclosed the character of the growth and subsequent careful physical examination of the testis revealed a suspicious nodule, or it was found after orchidectomy. All these hidden growths were chorionepitheliomata. Clinical diagnosis is based solely on the findings made by palpation except in the rare instances of a high content of hormone in the urine with negative findings on palpation. Ordinarily the testicle and its epididymis are distinguishable. In small growths an area of induration can be made out, but in larger growths the general shape of the testicle, as a rule, is maintained as in gumma. The examiner receives the impression, however, of an increase in weight or perhaps more nearly an increase in the specific gravity of the enlarged testicle. Even in advanced growths the epididymis can still be separated from the mass and difficulties in differential diagnosis usually arise only when this becomes impossible, or when gumma is present, or when the condition is complicated by coexistant hydrocele, epididymitis, etc. Care- ful search for metastases should always be made. The earliest site of metastasis of tumor of the testicle is the preaortic lymph zone but the time of occurrence is so uncertain that this fact has been abandoned as an indication for radical operation (orchidectomy with complete removal of the primary lymph zone of that side). Clinically it may be found that cells have crossed to the opposite lymph zone without giving evidence of their presence in the primary one. In one patient recently a large tumor developed on the right side months after a left orchidectomy for tumor of the testis. Supraclavicular glands on the left appeared a year after the removal of this retroperitoneal metastasis, at which time no masses could be felt on the other (left) side of the aorta and vena cava. In fact, metastases to the supraclavicular gioup of glands called "Virchow's nodes" are not at all infrequent with tumors of the testis. Metastases also occur to the viscera by way of the blood vessels. It is of course important to determine at the time of examination the presence or absence of any of these types of metastasis as well as their extent. The treatment of tumors of the testis has narrowed itself to simple orchidec- Frank Hinman 243 tomy and X-ray therapy. Definite diagnosis, of course, can be made only by the microscope and there is a distinct advantage therefore in early orchidec- tomy. I can see no benefit to be gained by preoperative irradiation which carries a risk of sterilization even with the best of lead protection for the good testis. W^ith high exposure of the cord at the external ring and its double liga- tion and division before manipulation of the tumor, orchidectomy w'ould seem to be perfectly free from any danger of spreading metastases. Ordinarily it is advisable to have collected urine for 24 hours or more for determination of the gonadotrophic hormone. There is no advantage in waiting for the result of this test, however, before performing orchidectomy inasmuch as neither the presence nor the absence of the hormone would modify the plan of treatment. The specimen removed is sent to the pathologist for careful serial examination and this study together with the report from the biochemi- cal laboratory gives reliable information as to prognosis. To those patients in whom metastases were not demonstrated clinically before operation, a prophylactic course of X-ray therapy to the preaortic lymph zone is given. In those patients in whom metastases were demonstrated before orchidectomy the areas of metastasis, as well as the preaortic lymph zone even if uninvolved clinically, are irradiated. Experience is not yet sufficient to permit any re- liable prediction of the effect of irradiation. As a rule the more primitive types of tumor are the most radiosensitive, that is, such masses disappear most readily under irradiation so far as clinical evidence goes, yet few of these patients are cured. On the other hand, the embryonal carcinomata, which are not nearly so primitive or radiosensitive, show the most favorable results after irradiation. During the first year it has been our custom to submit for biochemical assay 24- or 48-hour collections of urine from patients upon whom orchidectomy has been performed and followed by radiation therapy; thereafter such specimens are submitted at intervals of six months for the next two years. This has been done as a routine even on those patients who showed no increase of the hormone above normal before orchidectomy. Only on very rare occasions have abnormal amounts of the hormone appeared in these subsequent tests. Under the foregoing plan of treatment the prognosis depends on early recog- nition of the tumor, the pathologic type of growth found after orchidectomy, the finding of a functioning or nonfunctioning growth, and the presence or absence of metastases as determined by clinical examination. A good prognosis may be offered for patients with an early growth and no metastases clinically, with no gonadotrophic hormone in the urine, and with a type of tumor having fairly well differentiated cells. On the other hand, in spite of early diagnosis, a poor prognosis is all that can be advanced for the patient who shows a high content of hormone in the urine and a type of tumor having primitive cells even though no clinical evidence of metastases is present. The chorionepi- theliomata are the most malignant of all these tumors and, in our experience, always are fatal. Glancing back at table 1, it is seen in fact that no patient who 244 Tumors of the Testis showed gonadotrophic hormone in the urine before orchidectomy survived a year afterward in spite of the most vigorous irradiation. To summarize, patients without metastases and with no hormonal output have a good prognosis. For patients with metastases but still with no hormonal output the prognosis may be poor, fair, or even good, depending on the un- certain reaction of the particular tumor to irradiation. Finally, irrespective of metastases, patients who excrete gonadotrophic hormone have a very poor prognosis. HISTORY OF HYPOPHYSIAL DIABETES By B. A. HOUSSAY, M.D. From the INSTITUTO DE FISIOLOGIA DE LA FACULTAD DE CIENCIAS MEDICAS DE BUENOS AIRES, ARGENTINA HISTORY OF HYPOHPYSIAL DIABETES THE EVOLUTION of knowledge on the relation of the hypophysis to diabetes can be divided chronologically into five periods: ist, before 1924, the inci- dence of diabetes in acromegaly and the action of the posterior lobe; 2d, after 1924, the hypersensitivity of hypophysectomized animals to insulin; 3d, after 1929, the discovery of the action of the anterior lobe on diabetes; 4th, mis- conceptions prevailing in 1936; 5th, permanent diabetes induced by antero- hypophysial extracts. 1. Before 1924: the Incidence of Diabetes in Acromegaly and the Action of the Posterior Lobe. Since acromegaly was first studied, its coexistence with diabetic symptoms has been observed. Two papers collecting published cases may be mentioned on the subject: Xumber of cases Glycosuria Cases Per cent Borchardt' Atkinson- 176 817 71 . 268 40 12.8 The high incidence of diabetes in acromegaly prompted Borchardt^ to inject pituitary extracts; he found that they produced glycosinia and blood sugar increase of short duration. This effect is due to the posterior lobe and has been extensively studied by Houssay and Di Benedetto.^ In that paper many refer- ences to previous work are mentioned, besides the authors' experiments. In 1923 Burn^ discovered the antagonistic action on blood sugar of posterior- lobe extracts and insulin in the rabbit. This effect was also found in the dog by Magenta and Biasotti' as were several other antagonistic actions of these sub- stances (Houssay"). From then until 1929 the prevailing idea was that if the hypophysis had any action on carbohydrate metabolism and on the diabetes in acromegaly it was due to its posterior lobe. Gushing and his associates since 1911 have insistently studied the role of the hypophysis in carbohydrate metabolism. In the papers of Goetsch, Gush- ing and Jacobson" and of Weed, Gushing and Jacobson^ it is mentioned that in hypophysectomized dogs there occurs an increase in tolerance to sugar, since in these dogs the ingestion of large amounts of saccharose was not followed by glycosmia.* They also observed an increase in tolerance in hypophysectomized dogs in which a large part of the pancreas had been removed but did not ex- periment with totally pancreatectomized animals. These changes were attrib- * These results might be due to the slow absorption of sugar, as has been found in hypoph- ysectomized rats, and do not prove that more sugar is consumed. On the contrary, the injection of glucose into the veins of hvpophysectomized dogs produces a blood sugar increase which lasts longer than in normal dogs (Biasotti;'' Foglia and Potick^"). [ 247 ] 248 Hypophysial Diabetes uted to the absence of the posterior lobe because it was observed that they appeared in dogs in which that lobe had been removed and because the in- jection of posterior-lobe extracts decreases the sugar tolerance of normal and hypophysectomized dogs. Weed, Gushing and Jacobson" assumed that the hypophysis secreted a glyco- genolytic and glycosuric hormone which was liberated by the action of nervous stimuli sent by the superior cervical ganglion. But in 1930 Gushing" was in- clined to attribute these effects to the action of the hypothalamus: "Our find- ings may be capable of other interpretations now that attention has been drawn to the importance of the hypothalamic region whence possibly our glycogenolytic responses originated rather than in the hypophysis itself." 2. After 1924, the Hypersensitivity of Hypophysectomized Animals to In- sulin. After the discovery of insulin in 1922 it was prepared in Buenos Aires early in 1923 by Dr. Sordelli and we formed a team for studying its metabolic action and also its effects on the chemistry of blood and milk in different species, its antagonistic or coadjuvant action with different substances, and especially the influence of removing endocrine glands as well as the role of the nervous system, etc. A review on the latter subjects was published in Endocrinology in 1925 by Houssay.^" In 1923 the action of insulin in thyroidectomized (Ducheneau^^) and adre- nalectomized (Lewis and Magenta") animals was studied in our Institute. Houssay and Magenta^^ published in 1924 experiments showing the striking hypersensitivity of hypophysectomized dogs to insulin. These studies were extended in 1927^* and in 1929." In 1924 the great sensitivity to insulin was found in the toad after removal of the entire hypophysis or of the anterior lobe (Houssay, Mazzocco and Rietti'*). Three years afterwards (1927) Ceiling, Campbell and Ishikawa,^" confirmed our findings that hypophysectomized dogs are more sensitive to insulin but wrongly attributed the hypersensitivity to the absence of the posterior lobe. In 1927 Davidoff and Gushing"* wrote: "An experimentally induced hyper- glycemia, brought on by removal of a large portion of the pancreas, tended to subside when a partial hypophysectomy was subsequently performed." "If loss or diminution of the internal secretion of the pancreas robs the tissues of their power of metabolizing carbohydrates, certainly loss or diminution of the secretion of the hypophysial posterior lobe [sic] greatly enhances their power in this respect." Two important points are mentioned in this para- graph: 1st, the attenuation of pancreatic diabetes by hypophysectomy; 2d, the possibility that the part played by the hypophysis might not belong to its posterior lobe. Unfortunately the authors published no experimental details, such as the amount of pancreas removed, etc. It is well known that the changes produced by partial hypophysectomy or partial pancreatectomy are not per- manent and often subside spontaneously and completely. Moreover these studies were not properly completed by their authors and this is why they did not carry a general conviction. B. A. Houssay 249 In 1927 Davidoff and Gushing'" observed that insulin acts on the diabetes oi acromegaly but that larger amounts are needed to obtain normal blood sugar than in other types of diabetes.* They attributed the insulin resistance to a pituitary hyperactivity which would partially suppress the action of insulin and found that after removing a part of the adenoma the insulin resistance disappeared. They seemed inclined to assign some role to the anterior lobe: "Whether in acromegaly the melituria is due to an excess of posterior lobe extract, as the countereffect on insulin would appear experimentally to indi- cate, or to an overactivity of the acidophilic cells of the anterior lobe as we are inclined to believe, or possibly to a combination of both, for we know little of the source of the active substance of the posterior lobe, time will surely tell." In 1927 Johns, O'Mulvenny, Potts and Laughton^ observed that extracts of anterior lobe produced slight hyperglycemia and glycosuria. The preparation of the extracts was not published in detail but it was stated that they were free from protein. They may have erred in the last point, because according to our experiments the diabetogenic activity is contained in the protein frac- tion of the extracts. 3. After ip2p, the Discovery of the Action of the Anterior Lobe on Diabetes. Extending their studies on the hypersensitivity of hypophysectomized animals to insulin, Houssay and Potick" in 1929 confirmed the fact that the hyper- sensitivity appears in toads with no anterior lobe just as in those where the whole hypophysis had been removed. They also discovered that the implanta- tion of anterior lobe from toads or mammals (ox) protects hypophysectomized toads against the action of insulin. This finding led us to study with Dr. Biasotti the influence of hypophysectomy in pancreatectomized dogs and toads. I clearly recall the impression we received when we found that our first hypophysectomized dog did not excrete glucose in the urine during the three days following total pancreatectomy. These results w^ere first published in 1929 by Houssay" in a footnote of a paper in the jubilee volume dedicated to Prof. G. Araoz Alfaro. In 1930 Houssay and Biasotti'""'^ published their experiments in detail in three papers. The total removal of the pituitary in dogs'^ or toads^ or of the anterior lobe in the toad"^ produced a great decrease in the intensity of pancre- atic diabetes, also of phlorizin diabetes in the dog."' Implantation or injection of anterior lobe (posterior lobe was much less active) to hypophysectomized- pancreatectomized toads produced severe diabetes.^ These experiments allowed us to deduce the following conclusions: (1) the anterior hypophysis has a diabetogenic action; (2) its action on carbohydrate metabolism is physiologic and constant; (3) there exists a certain physiologic antagonism between the anterior hypophysis and the endocrine pancreas * In the extensive literature this resistance has been found present in some cases of acro- megaly but lacking in others. In the light of our present knowledge it may be assumed that in the insulin-resistant cases there actually exists an increased secretion of the anterior hy- pophysis, whereas in the nonresistant cases the pancreas had been damaged by an increased secretion of the anterior hvpophvsis which afterwards had sulisided and had become normal. 250 Hypophysial Diabetes (attenuation of pancreatic diabetes by hypophysectomy, hypersensitivity to insulin of hypophyscctoniized animals with or without pancreas, intense dia- betogenic action of the anterior lobe in absence of the pancreas); (4) fasting produces marked changes in hypophysectomized animals (rapid hypoglycemia and death in animals with or without pancreas, great decrease in glycosuria and in blood-sugar level of hypophysectomized-pancreatectomized dogs);^" (5) hypophysectomized dogs utilize food protein well but the endogenous ca- tabolism is decreased during fasting or in pancreatic and phlorizin diabetes;^' ^^ (6) hypophysectomized-pancreatectomized dogs utilize glucose, excreting only a part of the amount injected; hypoglycemia can appear during fasting, re- covery following rapidly on administration of glucose;""'^ (7) hypophysectomy decreases the ketonuria of pancreatic or phlorizin diabetes, as Rietti"" found in 1931 ; (8) the removal of the posterior lobe* does not alleviate the symptoms of pancreatic or phlorizin diabetes and does not increase the sensitivity to insulin in the dog. In the following years we found^^"'" that: 1st, the diabetogenic action is specific for the hypophysis, other organs being inactive; 2d, the anterior lobe of vertebrates (fishes, batrachians, reptiles, birds, and mammals) is active when injected into batrachians or mammals; 3d, removal of the hypophysis decreases the intensity of pancreatic diabetes in mammals, reptiles, batrachians, and fishes, and of phlorizin diabetes in mammals and batrachians. Oiu- work remained unnoticed from 1929 to 1931 and therefore we sent a summary of it to two of the leading American medical journals. The paper was not accepted by either of these journals but was finally published in Endocrinology.^ After the appearance of this paper two workers in the Physi- ological Institute of Chicago, Barnes and Regan,^ confirmed and extended our experiments. Since 1934 Long and Lukens*" '^ in important papers confirmed our results, using cats. In 1932 the diabetogenic action of the anterior hypophysis in mammals was observed. The three papers which appeared were in chronological order: 1, Evans, Meyer, Simpson and Reichert;"*^ 2, Bauman and Marine;'" 3, Houssay, Biasotti and Rietti (with Di Benedetto).^^ Confirmatory experiments were published in 1933 by Barnes and Regan," and E. I. Evans.^^ H. M. Evans and his co-workers"' *^ in their studies on the growth hormone of the hypophysis injected anterior-lobe extracts free from gonadotrophic sub- stances. They observed that a male dog after about eight months of continuous injection "developed marked skin infections and abscesses, became emaciated in spite of a large appetite, was easily exhausted, drank great amounts of water, and excreted great amounts of urine. This led to investigation of the urine of the injected male for the presence of sugar. It was present in large amounts. Fasting blood sugar was also high (232 mg. per 100 cc). The animal was ob- viously failing and injections were stopped to prevent its death. Four months * The alterations produced by dienceiDhalic lesions and total remo\al of the neinohy- pophysis are still not clearly understood (see Davis and co-workers,^" and Houssay=»). B. A. Hoiissay 251 after cessation of treatment, the mine still gave a positive Fehling test. The amount of sugar excreted had decreased and the urine vokime was less. The animal was more active and had gained weight." In the other group, one female dog injected for nine months had infection on the skin, hyperglycemia (228 mg. per loo cc.) and glycosuria; when the injections were interrupted the bitch remained glycosuric for two months and no abnormalities were found in its Langerhans islets. One female of the first group and a brother of the female of the second gi'oup which were treated likewise did not show dis- turbances of carbohydrate metabolism (blood and urine) even a year after being injected. Prof. Evans wrote to me about his experiments before publish- ing them and I mentioned this letter in my paper of 1932 (see Houssay'"). In the same year, Bauman and Marine'" injected four rabbits during six or more days with a saline extract of fresh bovine anterior lobe. The animals presented glycosuria, polyuria and an increase in blood sugar. The body weight did not decrease. Using the growth extract "Phyone," glycosuria ap- peared in one out of four rabbits. From 1932 up to now we have confirmed the Evans phenomenon several times, dogs injected daily with anterior lobe developing diabetes for some weeks during the treatment, remaining diabetic for one to four weeks after the injections were stopped, and then gradually becoming normal. The return of the blood sugar to normal is probably due to the disappearance of the lesions of the beta cells of the Langerhans islands. In 1930 Houssay and Biasotti^ had discovered the diabetogenic action of the anterior lobe by injecting toads deprived of hypophysis and pancreas. In 1932 it was confirmed in dogs without hypophysis and pancreas by Houssay, Bia- sotti and Rietti^^"'" who published three notes about it, one of them with Di Benedetto.''^ In these notes is also described the diabetogenic action of the anterohypophysial extract in: 1st, dogs with pancreatic mass reduced but nor- moglycemic (this particular point was later studied by Houssay, Foglia and Smyth'* in 1941); 2d, dogs with Sandmeyer diabetes; 3d, normal dogs. In the same year (1932) it was proved in our Institute that the anterohypophysial extract aggravates the phlorizin diabetes in toads*® and dogs."" ^' Finally in 1937, Foglia, Gerschman, Marenzi, Muiioz and Rietti"" showed that when injected into totally pancreatectomized dogs it aggravated the diabetes, in- creased especially the ketonemia and ketonuria, decreased the alkaline reserve, and killed them in coma within one to three days. In 1933 Houssay, Biasotti and Rietti*" compared the sensitivity of several animal species to the diabetogenic anterohypophysial action. They studied several chemical properties of the extract and demonstrated that its action can be observed in dogs without gonads, thyroid, adrenal medulla, sympathetic chains, and splanchnic nerves. In the hypophysectomized and pancreatecto- mized toad Houssay and Biasotti^''^" had shown from 1931 to 1933 diabetogenic action in absence of the digestive apparatus, lungs, kidneys, testicles, ovaries, thyroid, or forebrain, interbrain and midbrain. The diabetogenic action is not observed in absence of the liver in the pan- 252 Hypophysial Diabetes createctomized and hypophysectomized toad, as shown by Campos, Curutchet and Lanari" in 1933 and Foglia.=' By progressively reducing the amount of liver left it was shown that when the loss was over 55 per cent the diabetogenic action diminishes (Foglia'"). In dogs with full anterohypophysial diabetes, hepatectomy produces a quick fall of the glucemia and within a few hours they show serious hypoglycemic reactions, which disappear momentarily with glucose. According to Soskin"^ the anterohypophysial diabetogenic action is due to an overproduction of glucose in the liver. From 1934 to 1936 appeared the important works of Long and Lukens'"" and in 1937 one by Long,'' studying the influence of hypophysectomy on pan- creatic diabetes and the action of anterohypophysial extract in animals with both operations. The above studies led them to investigate the influence of adrenalectomy on pancreatic diabetes. In this field they did splendid work; their last paper (Long, Katzin and Fry^) was in 1940. 4. Misconceptioyis Prevailing in 19^6. During my trip to the United States in 1936 I realized that two false notions were held: 1st, it was not possible to observe the diabetogenic action in intact animals; 2d, the anterohypophysis had no intrinsic action on the metabolism of the carbohydrates, its action being exerted by way of the corticoadrenal hormones. Bennett, as well as Collip, and also Long and Lukens, had observed diageto- genic action in animals without hypophysis and pancreas, but not in normals. In reporting further work done in Evans's laboratory, RusselP in 1938 stated: "In our own laboratory, in spite of initial successes, more extensive trials in normal dogs and rabbits were unsuccessfid also, except in a few instances which were considered doubtful because of attendant protein reactions in the animals." In spite of this general incredulity, our Institute went on continuously ob- taining and studying this diabetogenic action, which we have by now observed in hundreds of dogs. The reason for our success was explained to all the above- mentioned investigators and to Young (November, 1935) on the basis of the following procedure used in preparing our extracts; immediately after the slaughtering of the animals the hypophyses were placed in carbon-dioxide snow and the dissection and subsequent preparation of the extracts was always carried out near 0° C; the glands were kept frozen from then on; we realized that at room temperature the glands and the extracts lose their action in a few hours. Among all these investigators only Young followed our advice; that is why he was able to accomplish the outstanding investigations well known to all of us. When Young"' '^ in 1937 ^"^^ 193^ published his results the doubts vanished; it explains why many believe he discovered the diabetogenic action of antero- hypophysial extract, which really has been well known since 1930 and 1932. The idea that the diabetogenic action is mediated exclusively through the adrenals comes from the work of Long. However, Long himself was always rather cautious in his conclusions, and in the paper by Long, Katzin and Fry^" B. A. Houssay 253 (in 1940) it was stated that "not all the diabetogenic effect of the anterior pituitary is mediated by the adrenal cortex, but rather that a synergism exists between these two hormones." The proof of the diabetogenic action of antero- hypophysial extracts in animals without adrenals was presented in the toad by Houssay and Biasotti'""" in 1933 and 1936 and by Houssay and Leloir™ in 1935, and in adrenalectomized dogs with reduced pancreas, kept alive with adrenal extract (Houssay and Biasotti^" in 1938) or with desoxycorticosterone or only with sodium chloride (unpublished experiments™). While we do not discard the possible existence of the synergism between anterohypophysis and corticoadrenal hormones, in the latter experiments no such synergism was detectable. Concerning participation of the adrenals in anterohypophysial action we shall publish some other experimental results later. F,. Permanent Diabetes hiduced by Anterohypophysial Extracts. In 1932 Houssay, Biasotti and Rietti''''''" proved that the injection of anterohypophysial extract into normoglycemic dogs with reduced pancreatic mass produces a diabetogenic action. After the interruption of the injection the majority of the animals returned to a normoglycemic state, but some remained perma- nently diabetic until death. This was the first demonstration of permanent diabetes due to the action of anterohypophysial extract. In 1937 Young" reported the very important verification, namely, that this particular permanent diabetes can be produced in normal dogs. His remark- able work was duly appreciated and the doubts of some concerning the antero- hypophysial diabetogenic action definitely subsided. Summaries of these works were published by Young"^""^ in 1939 and 1940. A year before, in 1936, Houssay and Foglia'" demonstrated that the pancreas of dogs with hypophysial diabetes showed a decrease in its capacity to secrete insulin. This was confirmed in 1941 by Houssay, Foglia, Smyth, Rietti and Houssay"" who also showed that if the dog resists the diabetogenic action and remains normoglycemic, or in case his diabetes subsides, tJie pancreas in many instances has a normal (and sometimes supernormal) secretion of insulin. Richardson"" and Young"" have shown that during anterohypophysial dia- betes the beta cells show a disappearance of granules, vacuolization and pyknosis; in case the animal remains with a permanent diabetes the islets are found atrophic, hyalinized, with few or no beta cells. If the animal resists the extract or if his hyperglycemia subsides, recuperation of the beta cells can be observed, as shown by Porto." Actually the anterohypophysial extract produces two kinds of diabetes: (1) the appropriately designated anterohypophysial diabetes observed during the injection of the extract; its origin is due first to extrapancreatic factors for only afterwards are the islets injured and the secretion of insulin dimin- ished. There is in these animals characteristic increase in resistance to insulin; (2) permanent diabetes, which is pancreatic and should be called metahypo- physial diabetes, being due to the lack of active Langerhans islets; the antero- 254 Hypophysial Diabetes hypophysial extract damages the islets, but its continued action is unnecessary during the permanent diabetic state. This history of anterohypophysial diabetes stops at this point without discussing the important contributions of Long, Chambers, Collip, Young, Soskin, and other investigators, because they are more recent and the exact chronology is better known. The discoveries made by Professor Herbert M. Evans in several fields of the biological sciences, among them in the physiology of the hypophysis, are many and fundamental. I am happy to associate myself in the homage now dedicated to him by writing this exposition of some fundamental points in the history of anterohypophysial diabetes, which I know as actor or coeval. Professor Evans was the first to demonstrate the anterohypophysial diabeto- genic action in mammals, and discovered that this effect may persist for several weeks after the treatment is ended. Later his pupils, Bennett and Russell, made other valuable contributions. I take this opportunity to express to Professor Evans my greatest admiration and sincere friendship. REFERENCES 1. Borchardt, L.: Ztschr. f. klin. Med. 66:332, 1908. 2. Atkinson, F. R. B.: Endokrinologie 20:245, 1938. 3. Houssay, B. A., and Di Benedetto, E.: Rev. Soc. argentin. de biol. 9:300, 310, 1933; Cpts. rd. Soc. de biol. 1 14:793, 795, 1933. 4.Burn,J.H.:Jl. Physiol. 57:318, 1923. 5. Magenta, M. A., and Biasotti, A.: Rev. Asn. m^d. argentin, 36:60, 1923; Cpts. rd. Soc. de biol. 89: 1 125, 1923. 6. Houssay, B. A.: Endokrinologie 5:103, 1929; Libra de oro en homenaje al Dr. Gregorio Araoz Alfaro (Buenos Aires: 1929). 7. Goetsch, E.; Gushing, H., and Jacobson, C: Johns Hopkins Hosp.Bull. 22:165, 1911. 8. Weed, L. H.; Gushing, H., and Jacobson, G.: Johns Hopkins Hosp. Bull. 24:40, 1913. 9. Biasotti, A.: Rev. Soc. argentin. de biol. 10:124, 1934; Gpts. rd. Soc. de biol. 117:54, 1934. 10. Foglia, V. G., and Potick, D.: Rev. Soc. argentin. de biol. 17:289, 1941. 11. Gushing, H.: Lancet 219:119, 175, 1930. 12. Houssay, B. A.: Endocrinology 9:456, 1925. 13. Ducheneau, L.: Rev. Asn. med. argentin. 36:944, 1923; Gpts. rd. Soc. de biol. 90:248, 1924. 14. Lewis, J. T., and Magenta, M. A.: Rev. Asn. med. argentin. 37:370, 1924; Gpts. rd. Soc. de biol. 92:821, 1925. 15. Houssay, B. A., and Magenta, M. A.: Rev. Asn. med. argentin. 37:389, 1924; Gpts. rd. Soc. de biol. 92:822, 1925. 16. Houssay, B. A., and Magenta, M. A.: Rev. Soc. argentin. de biol. 3:217, 1927; Gpts. rd. Soc. de biol. 97:596, 1927. 17. Houssay, B. A., and Magenta, M. A.: Rev. Soc. argentin. de biol. 5:99, 1929; Gpts. rd. Soc. de biol. 102:429, 1929. 18. Houssay, B. A.; Mazzocco, P., and Rietti, G. T.: Rev. Soc. argentin. de biol. 1:231, 1925; Gpts. rd. Soc. de biol. 93:968, 1925. 19. Gelling, E. M. K.; Gampbell, D., and Ishikawa, J.: Jl. Pharmacol. & Exper. Therapeut. 31:247, 1927. 20. Davidoff, L. M., and Gushing, H.: Archs. Internal Med. 39:751, 1927. 21. Johns, W. S.; O'Mulvenny, T. O.; Potts, E. B., and Laughton, N. B.: Amer. Jl. Physiol. 80:100, 1927. B. A. Houssay 255 22. Houssav, B. A., and Potick, D.: Rev. Soc. argentin. de biol. 5:66, 1929; Cpts. rd. Soc. de biol. 101 1940, 1929. 23. Houssay, B. A., and Biasotti, A.: Rev. Soc. argentin. de biol. 6:8, 1930; Cpts. rd. Soc. de biol. 104:407, 1930; Pfliigers Arch. f. d. gsmt. Physiol. 227:239, 1931. 24. Houssay, B. A., and Biasotti, A.: Rev. Soc. argentin. de biol. 6:251, 1930; Cpts. rd. Soc. de biol. 105:121, 1930; Pfliigers Arch. f. d. gsmt. Physiol. 227:664, 1931. 25. Houssay, B. A., and Biasotti, A.: Rev. Soc. argentin. de biol. 6:326, 1930; Cpts. rd. Soc. de biol. 105:126, 1930; Pfliigers Arch. f. d. gsmt. Physiol. 227:657, 1931. 26. Houssay, B. A., and Biasotti, A.: Endocrinology 15:511, 1931- 27. Biasotti, A., and Houssay, B. A.: Jl. Physiol. 77:81, 1932. 28. Houssay, B. A.: New England Jl. Med. 214:913, 1936; Bol. Acad. nacn. de med. de Buenos Aires 1936:183. 29. Rietti, C. T.: Archs. Soc. de biol. de Montevideo supp. 2:322, 1931; Jl. Physiol. 77:92, 1932; Rev. Soc. argentin. de biol. 10:136, 1934; Cpts. rd. Soc. de biol. 117:57, 1934. 30. Davis, L.; Cleveland, D., and Ingram, W. R.: Archs. Neurol. & Psychiat. 33:592, 1935. 31. Houssay, B. A., and Biasotti, A.: Rev. Soc. argentin. de biol. 7:3, 1931; Cpts. rd. Soc. de biol. 107:733, 1931. 32. Houssay, B. A., and Biasotti, A.: Rev. Soc. argentin. de biol. 9:29, 1933; Cpts. rd. Soc. de biol. 113:469, 1933- 33. Houssay, B. A., and Biasotti, A.: Bol. Acad. nacn. de med. de Buenos Aires 1933:64. 34. Houssay, B. A., and Biasotti, A.: Rev. Soc. argentin. de biol. 12:104, 1936; Cpts. rd. Soc. debiol. 123:497, 1936. 35. Houssay, B. A., and Biasotti, A.: Rev. Soc. argentin. de biol. 12:185, 1936; Cpts. rd. Soc. de biol. 123:824, 1936. 36. Houssay, B. A., and Biasotti, A.: Rev. Soc. argentin. de biol. 14:308, 1938; Cpts. rd. Soc. de biol. 129:1261, 1938. 37. Hotissay, B. A.; Biasotti, A.; Di Benedetto, E., and Rietti, C. T.: Rev. Soc. argentin. de biol. 8:570, 1932; Cpts. rd. Soc. de biol. 1 12:497, 1933. 38. Houssay, B. A.; Biasotti, A.; Di Benedetto, E., and Rietti, C. T.: Rev. Soc. argentin. de biol. 8:563, 1932; Cpts. rd. Soc. de biol. 1 12:494, 1933. 39. Houssay, B. A.; Biasotti, A., and Rietti, C. T.: Bol. Acad. nacn. de med. de Buenos Aires 1932:171; Rev. Soc. argentin. de biol. 8:469, 1932; Cpts. rd. Soc. de biol. 111:479, 1932. 40. Houssay, B. A.; Biasotti, A., and Rietti, C. T.: Rev. Soc. argentin. de biol. 9:489, 1933; Cpts. rd. Soc. de biol. 1 15:323, 325, 327, 1934. 41. Barnes, B. O., and Regan, J. F.: Endocrinology 22:193, 1933. 42. Long, C. N. H., and Lukens, F. D. W.: Science (n.s.) 79:569, 1934; Proc. Soc. Exper. Biol. & Med. 32:326, 1934; Proc. Soc. Exper. Biol. & Med. 32:743, 1935. 43. Long, C. N. H., and Lukens, F. D. W.: Jl. Exper. Med. 63:465, 1936. 44. Evans, H. M.; Meyer, K.; Simpson, M. E., and Reichert, F. L.: Proc. Soc. Exper. Biol. & Med. 29:857, 1931. 45. Evans, H. NL; Simpson, M. E.; Meyer, K., and Reichert, F. L.: Mems. Univ. California 11:425, 1933. 46. Bauman, E., and Marine, D.: Proc. Soc. Exper. Biol. & Med. 29:1220, 1932. 47. Evans, E. L: Proc. Soc. Exper. Biol. & Med. 30:1370, 1933. 48. Houssay, B. A.; Foglia, V. C, and Smyth, F. S.: Jl. Exper. Med. 74:283, 1941. 49. Di Benedetto, E.: Rev. Soc. argentin. de biol. 7:196, 1931; Cpts. rd. Soc. de biol. 107:1193, 1931- 50. Foglia, V. G.; Gerschman, R.; Marenzi, A. D.; Muiioz, J. M., and Rietti, C. T.: Rev. Soc. argentin. de biol. 13:83, 1937; Cpts. rd. Soc. de biol. 126:152, 1937. 51. Campos, C. A.; Curutchet, J. L., and Lanari, A.: Rev. Soc. argentin. de biol. 9:11, 1933; Cpts. red. Soc. de biol. 1 13:467, 1933. 52. Foglia, V. G.: Unpublished data. 53. Soskin, S.: Physiol. Revs. 21:140, 1941. 256 Hypophysial Diabetes 54. Long, C. N. H.: Harvey Lects. 32:194, 1937. 55. Long, C. N. H.; Katzin, B., and Fry, E. G.: Endocrinology 26:309, 1940. 56. Russell, J. A.: Physiol. Revs. 18:1, 1938. 57. Young, F. G.: Lancet 233:372, 1937. 58. Young, F. G.: Biochem. Jl. 32:524, 1521, 1938. 59. Houssay, B. A., and Leloir, L. F.: Rev. Soc. argentin. de biol. 11:464, 1935; Cpts. rd. Soc. de biol. 120:670, 1935. 60. Houssay, B. A., and Foglia, V. G.: Unpublished data. 61. Young, F. G.: Jl. Endocrinol. 1:339, '939- 62. Young, F. G.: New England Jl. Med. 221:635, 1940. 63. Young, F. G.: Endocrinology 26:345, 1940. 64. Houssay, B. A., and Foglia, V. G.: Rev. Soc. argentin. de biol. 12:237, 1936; Cpts. rd. Soc. de biol. 123:824, 1936. 65. Houssay, B. A.; Foglia, V. G.; Smyth, F. S.; Rietti, C. T., and Houssay, A. B.: Acts, y trabs., II Congr. pan-amer. de endocrinol. 1:222, 1941; Rev. Soc. argentin. de biol. 17:301, 1941; Fl. exper. med. 75:547, 1942. 66. Richardson, K. C.: Proc. Roy. Soc. London, s.B 128:153, 1940. 67. Porto, J.: Rev. Soc. argentin. de biol. 17:351, 367, 1941. THE SYMBALLOPHONE: A DOUBLE STETHOSCOPE FOR THE COMPARISON AND LATERALIZATION OF SOUND By WM. J. KERR X From the DIVISION OF MEDICINE LMVERSITV OF CALIFORNIA MEDICAL SCHOOL SAN FRANCISCO, CALIFORNIA THE SYMBALLOPHOxNE: A DOUBLE STETHOSCOPE FOR THE COMPARISON AND LATERALIZATION OF SOUND PAIRED ORGANS of hearing are a convenience and almost a necessity in the forms of animal life in which they are found. In the primitive sense, these animals are able to locate sources of sound which may indicate danger or a supply of food. The eyes are then directed to the point of origin of the sound and the identity of the agent may be determined. The hunter, whether animal or man, is greatly aided by the sounds produced by game in motion. Con- versely, the hunted instinctively recognizes the safety value of a fixed position with respect to both sound and sight. In our highly mechanistic age the man on the street is at a great disadvantage if he cannot locate an onrushing vehicle. The localizing functions of the paired organs of hearing are extremely sensi- tive. Psychologically, sounds arising from a point in the midline or in the plane of projection of the long axis of the head reach the tympanic membranes of both ears simultaneously and are of equal intensity. Unless the subject moves his head, he is unable to locate the origin of the sound in this plane which is perpendicvdar to the axis between the two ears. If, however, the sound arises from one side of the midline, the point of its origin may be lateralized with remarkable accuracy chiefly by means of the difference in the time of arrival and in intensity of the sound waves that reach the tympanic mem- branes of the two ears. The difference in the time of arrival of the sound waves is determined by the difference in distance between the source of the sound and the two ears; the difference in intensity is determined by the reduction in intensity caused by deflection of the sound waves which pass around the head to reach the ear on the side more remote from the source of the sound (fig. i). The simple device shown in figiue 2 readily demonstrates that a normal per- son's ears are capable of detecting differences in time of 0.000003 sec. This function may be tested by connecting a rigid hollow metal or glass tube of con- venient bore with flexible rubber tubing of equal bore and length, to the head- and earpieces of an ordinary binaural stethoscope. Small holes are drilled into the rigid tube, one exactly in the center, one on each side 5 mm. from the central aperture, and others equidistant from the central aperture but at variable distances from the center. The subject inserts the earpieces into his external auditory canals in the manner ordinarily employed by the physician and closes his eyes. Sounds are produced by blowing or by vibrating a tuning fork over the apertures. In normal subjects the sounds that enter the central aperture are not lateralized but those that enter either aperture 5 mm. from the center are readily lateralized to the side on which they originated. Sounds L 259 ] 26o The Symballophone that enter the apertures more remote from the center appear to arise at a greater distance from the midline. But when the difference in distance becomes as great as 15 to 18 cm., the sounds appear to arise almost opposite one ear. Sound travels in air at the rate of 330 m. per sec. If the ears are able to lateralize the sound that arises only 5 mm. from the midline (which under Ridht experimental conditions would ~ mean a difference in distance of 10 mm., or 1 cm., between the two ears) they must be ca- pable of detecting differences in time of only 0.000003 see- the time required for sound to travel 1 cm. in air. This time factor may be shown by the fol- lowing analysis: sound travels in air at the rate of 330 m. per sec, or 0.330 m. (330 cm.) in 0.001 sec, 0.0003 m. (0.3 cm.) in 0.000001 sec, and 1 cm. in 0.000003 sec. Some years ago, while we were studying the lateralizing and differentiating functions of the ears, one of my associates (Dr. A. M. Bassett) and I were paying the usual tribute to a Middle Western city between trains at a mov- ing-picture theater. During the interlude a trumpeter with three trumpets of different keys appeared on the stage. The trumpets were arranged as shown in figure 3. We were sitting in the balcony approxi- mately 30 m. from the trump- eter. When the trumpets were blown separately or in unison, we could differentiate accu- rately the sound produced by each instrument. Anyone with normal hearing can readily lateralize the sounds produced by instruments in an orchestra or a band or by voices of certain pitch or quality in a chorus. Furthermore, sounds such as voices may be differentiated and located accurately, for instance around the dinner table, at a tea or at a cocktail party. The organ of Corti is remarkably selective Fig. 1. When sounds arise from a point C which lies in the longitudinal plane of the head and is equidistant from the two ears, the sound waves reach the auditory nerves simultaneously. When sounds arise from any points to either side of point C, they arrive at the two ears after different intervals of time and always later in the ear more distant from the source of the sound, as shown by the broken lines. Likewise, the sound waves which travel through the greater distance are deflected in their course around the head and the intensity of the de- flected sound is reduced. The differences in timing and intensity of the sounds that reach the two ears are the two most important conditions in determin- ing the perception of lateralization. Sounds of differ- ent qualities from different external sources may be located simultaneously because of the highly selec- tive functions of the organ of Corti. Wm. J. Kerr 261 in detecting differences in quality of sounds and in lateralizing them through differences in timing, intensity, and other less important factors. When the hearing in one ear is defective, the ability to lateralize extraneous sounds may not be lost entirely if any sounds can be heard by the affected ear, since the ability to detect differences in timing may be retained. The stethoscope first described by Laennec in 1819 still is used by many European physicians. It consisted of a hollow wooden tube for mediate auscul- tation. It was interposed between the ear of the physician and the body of the Fig. 2. A simple device to demonstrate the remarkablv sensitive capacity of the ears to de- tect differences of time of 0.000003 {one three-millionth) sec. consists of a rigid hollow tube connected by rubber tubing of equal length and bore with the familiar headset of a stetho- scope. Apertures are made in the hollow tube at the central point C and at points R^ and L^ 5 mm. to the right and to the left of the central aperture. Other apertures may be made at any points R- and L- and R^ and L'' to the right and left of the central aperture. To test the lateralizing sensitivity of the two ears through function of timing, sounds are made directly over one of the apertines. When the normal subject listens while sounds are produced over the central aperture, he hears the sounds simultaneously in both ears; hence there is no lateralizing effect. When sounds are produced over apertures R^ or L'^ they will appear to arise at points to the right or left of the midline. Since these points are only 5 mm. removed from the center and the difference in distance to the two ears is 10 mm. or 1 cm., it is demon- strated that differences of 0.000003 sec. can be detected. If sound waves in air travel at the rate of 330 m. (i,o8o ft.) per sec, they \\ould travel 1 cm. in 0.000003 ^^c. Sounds that enter the device at points R^ and L" and R" and L^ appear to arise at points progressively more remote from the midline. The greatest differences in distance from the point of origin of sound to the two ears necessary to give complete lateralization would be about 15 to 18 cm., or one- fourth the circumference of the head. patient, and enhanced the accuracy of observations of sounds produced in the body, notably in the lungs and heart. Incidentally it protected as well the sensibilities of both the patient and the physician. Williams, in 1829, devel- oped the binaural stethoscope fitted with two earpieces and with two semi- rigid tubes joined to a device for application to the body of the patient. This 262 The Symballophone instrument was improved by Camman in 1855 and by Alison and many others in subsequent years. In use the original Laennec stethoscope was clumsy and the sounds heard through it appeared to arise directly in the ear. The binaural stethoscope constructed with connecting tubes of equal length and bore pro- jected the sound by illusion to a point directly in front of the physician. He in- stinctively turned his head toward the point to which the chest piece had been applied in order to ascertain the location of certain sounds. In making studies of homologous areas on the surface of the body, he had to move the chest piece between the two points to be compared. He could compare the qualities of different sounds but he could not perceive the finer variations. By this auscultatory method differences in timing of events in two areas that produced sound could not be detected, although when it was combined with palpation the move- ments of the thoracic cage and the events in the cardiac cycle could be correlated in some measure with the auditory evidence of mo- tion at nearby or remote points. During the century since the invention of the stethoscope, many improvements have been made in the chest pieces. Multiple stethoscopes have been devised that permit several auditors to listen simultaneously to a given sound. Audio-amplification has been adapted for use in large halls or by special wiring of individual receivers at- tached to the ordinary binaural stethoscope. A system of electrical filters has been introduced to peirnit study of the bands of frequency of sounds pro- duced in the body. In recent years several types of double stethoscopes have been described. Muralt^ in 1910 devised a stethoscope which permitted the physician to listen simultaneously to two areas over the lungs. This double stethoscope consisted of two chest pieces or acoustic bells each of which could be connected to one ear by a rubber tube. It is apparent from his illustrations that the length of the crossed tubes to the opposite ears was equal to that of the tubes passing directly from a chest piece to the ear on the same side. Sounds originating in either chest piece reached both ears simultaneously. More re- cently Froschels" suggested the use of a "differential stethoscope," which is essentially a pair of chest pieces connected to the ears by an X- (or four-way) connection to permit either an ipsolateral or a contralateral course of the sound to each ear. This instrument was devised to study the sounds produced in the vocal cords during phonation in cases of paralysis of the laryngeal nerves. Nicolai^ devised a "stereostethoscope" to observe differences in sounds originating in the two tempero-mandibular joints, and suggested its use in Fig. 3. Arrangement of the three trumpets described in the text. Wm. J. Kerr 263 other fields of clinical medicine. This instrument consisted of two chest pieces connected by two separate rubber tubes to their respective earpieces. Hantsch- mann and Nicolai' reported that this instrument was of value in the study of cardiac and pulmonary disease. In 1935 Hawthorne^ described a differential (double) stethoscope similar to that devised by Nicolai. Alison" had described this type of instrument in 1858 and had given a report on its use in 1859. My personal interest in the possible value of a comparing stethoscope was aroused in 1936 when before a meeting of the journal club of the Division of Medicine one of my colleagues, Dr. T. L. Althausen, gave a re- port on Nicolai's paper.* Dur- ing the discussion I pointed out that, in nature, sounds which are lateralized or com- pared are heard in both ears and that differences in timing and intensity are factors which aid in locating and comparing sounds. In Nicolai's instru- ment the sounds to be com- pared could not be heard in both ears under the conditions existing in nature. During the afternoon following this dis- cussion, rubber tubing, four Y- pieces, two similar chest pieces, and a headset from an ordi- nary binaural stethoscope were assembled into a clumsy but working model of the symbal- lophone. The crossed tubes passing from the two chest pieces to the contralateral ears were made longer than the di- rect tubes, that is, those passing from the chest pieces to the ipsolateral ears. Subsequently the bore of the crossed tubes was reduced^ (fig. 4). This crude working model was capable of giving information not readily obtainable by the binaural stethoscope.* Homologovis areas in the two lungs could be compared by a method that employed the normal functions of the ears for comparing sounds. Sounds produced in the circulatory system by contraction of the heart and by the movements of the pulse wave could be readily compared. In disease, mur- murs propagated on the pulse wave beyond the heart were heard as soinids in motion, in contrast to the isolated sounds heard with the binaural stetho- O O Fig. 4. (A) Nicolai's comparing stethoscope. (B) The original symballophone showing the arrangement of tubing. The diagonal tubes are longer than the lateral tubes, and the bore of the crossed tubes is reduced. The sounds to be compared are heard in both ears as would be the case with sounds heard in nature. (C) A possible variation of (B) but impractical because the connections ^vould permit sound waves to pass freely in either direction from the transverse connecting tube. 264 The Symballophone scope. While clinical observations were being made, experimental studies were begun to determine the most effective device for enhancing the illusion pro- duced by the use of tubes of varying length and bore. Through the cooperation of Professor C. W. Brown of the Department of Psychology, University of California, the studies which will be described were made. By means of a gravity-pressure system, water was caused to flow through rubber tubing arranged in such a way that the flow could be impeded by metal R«p7»ee«nt arcentpic pump. Fig. 5. — A device for the study of the capacity of the two ears to detect differ- ences in intensity and timing of sounds with the symballophone. Water under hydrostatic pressure was allowed to flow through the system of tubing at A or B. An eccentric pimip E gave an impetus to the current to represent the pulse wave and to accentuate the sounds produced by faults in streamlining at points ^1,2,3,4 gj^jj g 1,2, 3, 4 Metal cylinders of different bore were placed in the rubber tubing at points A^' -• *■ * equidistant from the point of inflow, and metal cylinders of equal bore were placed at points B^--^* at varying distances from the point of inflow. For comparison of sounds the two chest pieces of the symballophone were placed over the two obstructions where sounds were to be compared. In this manner differences in intensity of soimd were studied over A^-'''* and differences of timing over B.^-'^'* cylinders with bores (1) of varying diameters or (2) of equal diameter but so placed that they were at unequal distances from the point of inflow. The vibrations caused by the water passing through these narrowed apertures, which could be heard as sounds (murmurs), were studied with the aid of the symballophone. An eccentric pump was inserted in the pressure system above the obstructions to simulate the pulse wave and to enhance the sounds (mur- murs). Figure 5 shows the details of the arrangement.* Soon it was apparent that the ears are so sensitive that extremely small dif- ferences in intensity or time could be readily detected by untrained persons * Throughout this study I had the assistance of two able senior medical students, A. M. Bassett and M. J. Goldman, to whom I am greatly indebted. Wm. J. Kerr 265 as well as by those experienced in the use of the symballojilionc. Differences of 1 per cent in the flow through any two obstructing cylinders could be differ- entiated by observing differences in intensity of the sounds produced. It was almost impossible to arrange obstructions of equal bore, so that the smallest differences in time could not be noted. However, by varying the bore and length of the crossed sections of tubing, the differences in length and bore which gave the most satisfactory results were obtained. When the crossed tubes were approximately 15 cm. longer than the direct tubes, the opti- mal difference in timing to the two ears was secured. When the crossed tubes had an internal diameter approximately three- fourths that ol the direct tubes, the differences in intensity of the sound which reached the two ears could be made to stipplement the illusion of lateralization. A number of models with ex- changeable metallic parts of vary- ing length and bore were con- structed. By practical experience in normal and abnormal subjects, it was found that the length and bore of the tidies previously de- scribed were the most satisfactory. The final model" (fig. 6) differed from model B shown in figure 4 in that the crossed and direct tidies were made of rigid metal tubing to insure permanency of the essential functioning parts. The tubes from the chest pieces to the central tmit were made of rubber tubing of equal length and bore to permit flexibility in use. Rubber tubes of equal length and bore were inserted to connect the central unit to the earpieces. Strong lateral steel-band springs were fixed to exert equal pressure on the earpieces in the external auditory canals.* Clinical Observations Lungs. In examining the sounds produced in the lungs dining respiration (auscultation), the physician who uses the ordinary binaural stethoscope usu- ally moves the chest piece fiom one point to another in order to compare * The symballophone has been patented, and in order to protect it trom exploitation, tlie rights under patent \o. 2,209,164 have been assigned to the Regents of the University of Cali- fornia. The instrument is made and sold at or near cost with the iniderstanding that any profit which may accrue from its manufactme and sale will be used for medical research. Fig. 6. The s)niballophone as finally constructed al ter experimental and clinical studies. 266 The Symballophone homologous areas for differences in duration, pitch and intensity of sound and to discover and localize abnormal sounds. By this method he is unable to detect a lag or delay in motion on one side or in one area although the pal- patory method may be used to detect gross changes in timing. In comparing the qualities of the sounds in different areas, he must remember the sounds heard at one area while he is listening over another. By use of the symballo- Fig. 7. Points on the chest at which the two chest pieces of the symballophone may be ap- plied to compare homologous areas. In this manner the respiratory sounds may be compared more accurately on the two sides of the chest. The examination may be completed in approxi- mately one-third the time required when the ordinary stethoscope is employed. Two areas over either lung may be similarly compared. A, ventral view; B, dorsal view of chest. phone (fig. 6) two areas may be compared simultaneously. The auscultatory signs in the chest may be examined more quickly and accurately than by the use of the ordinary stethoscope. As shown in the diagram (fig. 7), homologous points may be explored uniformly. Differences in timing of the respiratory movements and differences in pitch and intensity can be readily noted. Locali- zation of abnormal sounds (ronchi, rales) can be determined through laterali- zation. Sounds of different quality may be differentiated in the two areas under observation. The physician with some training may observe a number of differ- ences in respiratory sounds during a single complete respiratory cycle. It is well known that the respiratory sounds over the right apex of the lung dorsally are harsh, but it is not so well known that the sounds over the left apex ventrally are harsher. These differences probably are due to the relative Wm. J. Kerr 267 differences in volume of the pulmonary tissue through which air is moving on the two sides, which are determined by the position of the aorta. The ascend- ing aorta encroaches on the right lung ventrad and the descending aorta on the left lung dorsad. By the use of the symballophone, areas of consolidation, collections of fluid or air, and other clinical conditions can be localized more accurately than by any other auscultatory method. Medical students and physicians alike have little difficulty in locating areas of abnormality in the chest because the device makes use of the normal functions of the ears for the comparison and laterali- zation of sound. The physician need not understand the details of construc- tion.* He becomes concerned with the explanation of the causes for the dif- ferences he observes in the areas of the chest. However, he must give some thought to the functions of the membranes and the airways of the lungs in which the sounds are produced if he is to interpret the significance of the abnormal sounds he hears. Heart and Blood Vessels. In early life the closure of the semilunar valves is accompanied by a sound which is of greater intensity or louder in the second left than in the right intercostal space equidistant from and adjacent to the costal margin. During and beyond middle life the intensity of the sounds over these two areas is reversed. In the first instance we say that the pulmonary closure (second) sound exceeds or is greater than the aortic closure (second) sound, and in the second instance, that the aortic closure (second) sound ex- ceeds or is greater than the pulmonary closure (second) sound. Many physi- cians carelessly place the single chest piece of a binaural stethoscope over the right and left second or third intercostal spaces and merely record the relative intensity of the second sounds. With the symballophone the classical differ- ences in the second intercostal space can be readily demonstrated. In the third intercostal spaces the second sounds over the pulmonary valve are almost always louder than over the aortic valve even when aortic hypertension exists. A great increase in the pulmonary second sound is observed in patients who have the clinical condition known as cor pulmonale. The second or closure sounds of the semilunar valves can readily be com- pared in intensity with the first or systolic sound heard at the apex which accompanies contraction of the ventricles. In myocardial failure this systolic sound at the apex frequently is diminished in intensity although it may be heard faintly in other conditions such as ptilmonary emphysema, pleural effu- sion, pneumothorax, or pericardial disease with an accumulation of fluid or air in the pericardium. Murmurs that arise in connection with defects in the valves travel with the pulse wave. Diastolic murmurs are not propagated beyond the heart since they are associated with the flow of blood into the cardiac chambers. Systolic murmurs that arise in the heart, however, are of two types and may be de- * An analogous situation exists when a person with normal stereoscopic vision uses Oliver Wendell Holmes's stereoscope. He perceives a sensation of depth without being required to know how the device is constructed. 268 The Symballophone scribed as (i) those which are propagated on the pulse wave and leave the heart through the aortic or pulmonary valve, and (2) those which are propa- gated through the tricuspid or mitral valve into the auricles. The former travel widely on the pulse wave along the course of the aorta or pulmonary artery and frequently are accompanied by a palpable thrill because of the proximity of these vessels to the surface of the chest. Those propagated along the pulmonary artery are not transmitted widely, in no case beyond the chest. Systolic murmurs associated with backflow through the tricuspid valve are heard over the lower portion of the sternum which is nearest to the right auricle. The systolic murmur of mitral insufficiency is heard best at the apex Direction o"f -flow. IJiito^ ilppv"^"^""' Murmurs Produced by Faults in Strf.amlimng Murmurs travel both ways from point of constriction, but for a greater distance in the direction the liquid is flowing. Fig. 8. A simple elastic tube such as a garden hose through \\hich liquid is flo\ving may be used to illustrate the effects of compression that cause faults in streamlining. Obstructions in the heart at the orifices of the valves, or variations in the size of the lumina of vessels, cause faults in streamlining which produce vibrations that may be heard as murmurs. and is transmitted toward the left axilla and left subscapular region which is nearest to the left ventricle and auricle. The intense systolic pressure in the left ventricle drives the blood backward through the aperture in the defective mitral valve with such force that the sounds produced are louder and more widely propagated. The proximity of the contracted left ventricular muscle in systole to the wall of the chest permits more direct propagation of the vibrations which produce the murmur. The vibrations travel in both direc- tions from the point of origin but for a greater distance in the direction of the blood flow, as will be described. If a simple compressible elastic tube, for example, a garden hose, through which liquid is flowing (figs. 8 and g) is compressed, vibrations are produced in the walls of the tube and eddies form in the liquid. If the tube is palpated in front of and beyond the point of constriction, the vibrations are felt as thrills. AV^ith the stethoscope, sounds are heard which may be designated as murmurs. These sounds may be heard for some distance in both directions from the point of compression but always for a greater distance in the direction of the flow of the liquid. Let us now apply these observations to the heart and blood vessels. Two good clinical examples of this kind are aortic stenosis and pul- monary stenosis in which the diseased valves provide the partial obstructions, Wm. J. Kerr 269 and the ventricles, aorta, and pulmonary artery provide the channel through which the blood is flowing during systole. The pulse w^ave travels through the aorta at the rate of about 15 m. per sec. If the carotid vessels or the aorta are palpated in the neck or through the chest over the upper sternum, a systolic thrill appears on the crest of the pulse wave. If one chest piece of the symballophone is applied over the aortic area (second right intercostal space near the sternum) and the other above the course of the carotid artery or even 1\/Iur>mur> ^pavelsl)'. vvitVi the pulse wave EddiGS in the blood slr^eam Coar-ciaiion of aoT>ia Aor-iic stenosis A^ Fig. 9. The transmission of the systolic uuuniin forward in the direction of the blood flow from the stenosed aortic vahe and backward along the walls of the left \entricle is shoAvn diagrammatically. At the point X in the descending aorta two types of congenital lesion are encountered. In each instance the timing of events depends upon the arrival of the pulse \va\ e at this point bet\\een 1/20 and 1/10 sec. following the first or svstolic sound at the apex. In coarctation of the aorta a systolic miuiinu' is heard over the left upper chest dorsad, and in patent ductus arteriosus (Botalli) the systolic accentuation of a continuous nrurmur is heard over the pulmonary artery ventrad, chiefly to the left of the sternimr and below the clavicle in the second and third intercostal spaces. over the sternum at a point cephelad to the other chest piece, the systolic mur- mur heard appears first at the point nearer the aortic valve and later at the more distant point. After a simple auditory demonstration of the sensory effect of blowing across the two chest pieces, the natural perception of a sound in motion can quickly be recognized. The sounds heard appear to be "alive" in contrast to the rushing noise heard with the ordinary stethoscope. The systolic murmur over the carotid arteries appears about 0.05 sec. after the first sound 2'7o The Symballophone at the apex which accompanies the systolic contraction of the ventricles. Differences in time of this magnitude are readily determined by the use of the symballophone in which the crossed tubes are 15 cm. longer than the direct tubes, since with it differences in time of only 0.000003 sec. can be detected. Furthermore, patients with aortic stenosis may present a systolic murmur at the apex as well as at the base of the heart. With the symballophone it may be demonstrated that the murmur appears at the apex after it has been heard at the base. Apparently this murmur is propagated along the muscle of the left ventricle when it is in a contracted state. With the ordinary stethoscope this systolic murmur sometimes is diagnosed as a midsystolic murmur. If the aortic valve is calcified and the murmur is very loud, the demonstration is easy even for the novice. See figure 9. In patients with lesions of the pulmonary valve, congenital or acquired, conditions similar to those described for the aortic valve exist. The systolic murmur is not widely propagated; it is heard best in the region of the pul- monary valve moving outward and upward toward the left clavicle; it is heard faintly in the second right intercostal space adjacent to the sternum where the right pulmonary artery passes behind the ascending aorta. This murmur is not propagated beyond the larger branches of the pulmonary artery. In patients who have had long-standing pulmonary stenosis, the systolic murmur is prolonged, probably because the thin-walled pulmonary artery beyond the obstruction is widely dilated. The time of maximum intensity appears to be during or soon after the first or systolic cardiac sound. Two and possibly three types of congenital heart disease lend themselves to analysis by means of the symballophone (fig. 9). In patent ductus arteriosus (Botalli) the murmur which usually is present is heard best in the region of the pulmonary artery beneath the left clavicle. Frequently it is a loud con- tinuous murmur with a churning or rushing sound in the systolic phase. If one chest piece of the symballophone is applied over the apex of the heart and the other over the pulmonary artery, it will be readily noted that the time of the systolic accentuation of the murmur is much later than the time of the first apical sound. This observation indicates that the pulse wave has traveled over a prolonged route which in this case is via the aorta to the ductus and thence to the pulmonary artery. In my experience these conditions have been met in no other clinical state. Recently, in a patient with a patent ductus arteriosus (Botalli) the character of the murmur changed after the ductus was tied off. The delay noted in the time of appearance of the systolic murmur over the pulmonary artery could be explained by the great dilatation of this thin-walled vessel after many years of strain from increased arterial pressure transmitted from the aorta through the ductus arteriosus (Botalli). Theoreti- cally, in a patient with patent interventricular septum the pulmonary artery may become similarly dilated and thus cause a systolic murmur from relative pulmonary stenosis through dilatation of the pulmonary artery after long- continued hypertension in the lesser circulation. Wm. J. Kerr 271 In coarctation of the aorta a systolic murmur frequently is heard over the upper thoracic spine and between this area and the left scapula. If one chest piece of the symballophone is placed over the apex and the other over the area in which the murmur is heard, it will be readily noted that the murmur appears appreciably later than the first or systolic sound at the apex. The reasons are that the area of coarctation usually is found in the descending limb of the thoracic aorta and that the pulse wave arrives in the aorta at that point about 0.05 sec. after the systolic contraction of the heart. Arteriovenous aneu- rysms in the orbit, cranium, and extremities lend themselves to more accurate study with the symballophone than with the ordinary stethoscopes. Blood- Foelal loGs-plpa^e: 160 per^ minute, IGbeais per^ Gseconda. Yoelal he3.v\ rale: ]60ipev minu-ie, UlDeais pep Gseconds. Fig. 10. Schematic representation of the auditory effects produced by two rhythmic sounds of different rates as heard over the symballophone. When the rates are almost identical they may be easily differentiated by the cyclic return of synchronous and asynchronous phases. Foetal heart sounds in twin pregnancy in which the heart of each foetus has its own cardiac rate may be taken as a clinical example. The foetal and maternal cardiac rates may be simi- larly compared. pressure measurements by the auscultatory method may be made by applying either chest piece of the symballophone over the artery distal to the point of compression. With the symballophone foetal heart sounds may be compared with mater- nal heart sounds. Such observations are of importance when the maternal cardiac rate is high and the question of a viable foetus arises. Differences of a single heartbeat per minute may be readily detected. Twin pregnancies may be diagnosed if the two chest pieces are placed over the areas in which distinct foetal sounds are heard. As the physician listens to the sounds, he will observe that at times the two sounds are asynchronous, then they gradually approach each other and finally they coincide. Thereafter they gradually move apart and again approach each other in the next cycle, as shown in figure 10. Unless this pattern of sound is kept in mind, the obstetrician will not be able to make an accurate comparison when there is little difference in the foetal heart rates. Vocal Cords. Vibrations of the vocal cords produce a variety of sounds through variations in length and in tension of the membranes. If one of the paired cords is diseased or contracts imperfectly because of weakness of the controlling muscles, abnormal vocal sounds are produced. In clinical medi- cine it frequently is important to know whether hoarseness is due to local disease or to paralysis of the muscles of phonation. If the physician stands directly behind the patient and applies the chest pieces of the symballophone Fig. 11. Position of the examiner while the sounds produced by the vocal cords are being compared. The drawing illustrates the method of placing the two chest pieces equidistant from the larynx. The right vocal cord in this example is paralyzed and does not vibrate normally. The nor- mal (left) cord produces a more intense sound than the defective (right) cord and the sound is lateralized to the normal (left) side. The examiner places himself in a similar position directly behind the patient when he examines the antra. The louder sounds are lateralized to the normal or unaffected side. Wm. J. Kerr 273 evenly to the sides of the larynx and at other points equidistant from the larynx, as shown in figme ii, the paralyzed vocal cord can be readily deter- mined. The sounds are of greater intensity on the side on which the vocal cord is intact while the paralyzed cord is incapable of producing a normal sound. The sound used for the determination is repetition of a high-pitched tone and may be produced by having the subject pronounce the long letters E or A. In a large number of patients who had paralysis of one vocal cord due to goiter with associated injury of the laryngeal nerves or due to aneurysm of the aorta with associated injury of the left recurrent laryngeal nerve, we have been able to discover the site of the paralysis. In some of these patients direct laryngeal examination of the vocal cords had given doubtful evidence of the paralysis. Ajitrum of Highmore. If one antrum of Highmore is filled with fluid, the nasal tones do not reverberate over it as well as they do over the unaffected sinus. The symballophone is useful in detecting this abnormal state. The physician stands directly behind the patient and while he applies the chest pieces evenly over the antra the patient repeats "ninety-nine," "nine unknown men," or some other combination of nasal tones. The louder sound is lateral- ized to the side of the normal antrum. Malingerers of Hearing may be discovered by use of the symballophone. In this instance the earpieces are inserted in the patient's ears. After he has been blindfolded, sounds are made by the physician over either chest piece. If the hearing in one ear is very defective, the sounds that arise in either chest piece are referred to the ear with normal hearing. By alternately clamping the rub- ber tubing in the headset of the instrument, sounds may be deflected to either ear. Thus it is impossible for the malingerer to escape detection. Joints and Muscles. By means of the symballophone, sounds in paired joints in motion may be compared. This method is particularly useful in the study of the tempero-mandibular joint. Uneven mastication may resiUt in injury to one joint. Faulty or uneven movements and abnormal sounds may be noted and lateralized. Differences in the time when events occur in the joint can be readily determined. In studying paired muscles with the symballophone, the impaired sound may be detected on the affected side by comparing the purring sounds produced when the muscles are stretched or contracted. This observation may be important when slight differences occur in the mus- cles of the two sides of the body, as in minor degrees of paralysis or in residual weakness on one side following hemiplegia or monoplegia which is too slight to be detected by ordinary means. Stomach and Intestine. Peristaltic sounds in the stomach and large bowel may be studied by the use of the symballophone. It may be important to de- termine whether peristaltic sounds in the epigastrium are moving from left to right or from right to left. The former would suggest that the sounds arose in the stomach; the latter, that they arose in the transverse colon. The symballophone is of no value to persons whose hearing is defective in 274 The Symballophone one ear. Among medical students, about lo per cent have auditory defects which preclude any practical use of the symballophone. To some of them the binaural stethoscope is of no more value than the original monaural stetho- scope would be. Older physicians whose auditory functions are failing do not derive additional information from the use of the symballophone. Summary The symballophone, an improved double stethoscope for lateralization and comparison of sounds, has been described. A brief historical sketch has been given of the evolution of the stethoscope, and previous types of lateralizing stethoscopes have been mentioned. The psychological and clinical studies which led to the perfection of the symballophone have been discussed. The application of the symballophone to medical practice has advanced our inter- est in the art of auscultation and our knowledge of the fundamental principles involved in the moving structures in which sounds are produced. The symbal- lophone should replace the ordinary binaural stethoscope for all clinical purposes. REFERENCES 1. Muralt, L. V.: Btrge. z. Klin. d. Tuberk. i6: 121, 1910. 2. Froschels, E.: Med. Klin. 30:1099, 1934. 3. Nicolai, L.: Klin. Wchnschr. 15:91, 1936. 4. Hantschmann, L., and Nicolai, L.: Klin Wchnschr. 15:92, 1936. 5. Hawthorne, C. O.: Irish Jl. Med. Sci. (7[i.e., 6]) 1935:49. 6. Alison, S. S.: Med. Times & Gaz. (n.s.) 19:7, 28, 1859. 7. Kerr, W. J.; Althausen, T. L.; Bassett, A. M., and Goldman, M. J.: Amer. Heart Jl. 14:594, 1937- 8. Kerr, W. J.; Althausen, T. L.; Bassett, A. M., and Goldman, M. J.: Trns. Assn. Amer-Physns. 52:92. 1937- 9. Kerr, W. J.: West. Jl. Surg., Obstet. & Gynecol. 49:632, 1941. ON THE SIGNIFICANCE OF THE FORGOTTEN THERMODYNAMIC THEOREMS OF CARNOT By F. O. KOENIG From the DEPARTMENT OF CHEMISTRY STANFORD UNIVERSITY, CALIFORNIA ON THE SIGNIFICANCE OF THE FORGOTTEN THERMODYNAMIC THEOREMS OF CARNOT I. Introduction EXAMINATION OF THE famous memoir by Nicolas-Leonard-Sadi Carnot (1796- 1832) entitled Reflexions siiv la Puissance Motrice du Feu et sur les Ma- chines propres a Developper celte Puissance (A Paris, chez Bachelier, libraire, Quai des Augustins, No. 55, 1824)* shows that this work, ahhough in the form of a single chapter without subheadings, falls naturally into three distinct parts. In the first part {Ca pp. 1-38, Ma pp. 3-21) Carnot describes the cycle and demonstrates the theorem which have both come to bear his name. As is well known, Carnot's demonstration consists of a logically correct argument from two premises, one of which is true and the other false: the true premise is the impossibility of perpetual motion, and the false one the caloric theory of heat, according to which heat is a weightless indestructible substance. Carnot's theorem is thus an outstanding example of a physically true statement which is rigorously deducible from premises which are at least partly false. In the second part (Ca pp. 39-88, Ma pp. 21-46) of the Reflexions, Carnot deduces seven further theorems relating mostly to the thermal behavior of gases, and illustrates these theorems to some extent by numerical data and calculations. It is these seven theorems that we here refer to as "the forgotten thermodynamic theorems of Carnot." The arguments by which Carnot arrives at these theorems are again logically correct and again based upon two prem- ises, one of which is true and the other false: the true premise here is Carnot's theorem and the false one again the caloric theory of heat. In the light of modern knowledge it is obvious that any theorem deduced in this way must belong to one of the following three classes: class A, theorems based on Car- not's theorem alone and therefore true; class B, true theorems based either on the caloric theory alone or on the caloric theory and Carnot's theorem simultaneously; class C, false theorems based either on the caloric theory alone or on the caloric theory and Carnot's theorem simultaneously. Class B would evidently furnish further examples of true statements rigorously deducible from premises at least partly false. In the third part {Ca pp. 89-1 18, Ma pp. 47-61) of his memoir Carnot com- pares in the light of the two preceding parts the practical advantages of differ- * The first edition of this work will here be referred to bv the abbreviation Ca. An English translation with a few notes by W. F. Magie is included, together with writings by Claiisius and by William Thomson, in volinnc VI of "Harper's Scientific Memoirs" (ed. by J. S. Ames; New York and London: 1899); this volume, which is entitled The Second Law of Thermo- dynamics, will here be referred to by the abbreviation Ma. An annotated German translation of the Reflexions by Wilhelm Ostwald, containing further bibliographic information on Car- not, constitutes Nr. 37 of "Ostwald's Klassiker der Kxakteu Wissenschaften" (Leipzig: 1892). c -^77 : 278 Thermodynamic Theorems of Carnot ent types of working substances and points out how far the efficiency of steam engines actually in operation in his day falls short of the theoretical maximum. Of these three parts of the Reflexions, the third is a contribution primarily to technology, while the first and second contain Carnot's contributions to science. Of these contributions to science, in turn, by far the most important are contained in the first part of the Reflexions: Carnot's cycle and Carnot's theorem on reversible engines are the ideas which in the minds of Clausius (1850) and of William Thomson (1851) gave birth to the second law of thermo- dynamics. It is moreover only this outstanding part of his achievement by virtue of which Carnot lives in the consciousness of scholars today; the exist- ence of the second part of the Reflexions appears to have been completely forgotten. Our object in this article is to show that Carnot's forgotten theorems deserve some attention both from historians of science and from scientists of the present. II. Critical Summary of the Forgotten Thermodynamic Theorems These theorems, quoted from Ma in the order of their enunciation and num- bered by us in that order, are as follows: 1. When a gas passes without change of temperature from one definite vol- ume and pressure to another, the quantity of caloric absorbed or emitted is always the same, irrespective of the nature of the gas chosen as the subject of the experiment. 2. [At a given pressure] the difference between the specific heat [referred to unit volume] under constant pressure and the specific heat at cojistant volume is the same for all gases. 3. When a gas changes in volume without change of temperature the quanti- ties of heat which it absorbs or gives up are in arithmetical progression ichen the increments or reductions of volume are in geometrical progression. 4. When the volume of a gas increases in geometrical progression its specific heat [referred to unit zveight] increases in arithmetical progressioji. 5. [For a given gas at a given temperature] the difference betiueen the specific heat under constant pressure and that at constant volume is always the same, whatever the density of the gas, provided the quantity of gas by iveight remaijis the same. 6. The quantity of heat due to the change in volume of a gas [at constant temperature] becomes greater as its temperature is raised. 7. The descent of caloric produces more motive power at loiver degrees of temperature than at higher. Comments. (1) Gases are assumed throughout to obey the laws of Boyle and Gay-Lussac, that is, to be perfect. (2) All the proofs given in the text proper are of the synthetic quasi-Euclidean type, but in a long footnote (beginning on Ca p. 73, Ma p. 39) Carnot gives alternative analytic proofs of theorems F. O. Koenig 279 1, 3, 4, 7. (3) Scrutiny of Carnot's proofs in the light of our present knowledge reveals the fact that the distribution of the theorems among the three neces- sary classes mentioned above is as follows: Class A: theorems 1, 3 Class B: theorems 2, 5, 6, 7 Class C: theorem 4. The proof of this distribution would of course require extensive quotation from the Reflexions and must be omitted here for lack of space. III. Significance for the History of Science The forgotten theorems of Carnot are the first examples in thermodynamics of special results of physical interest obtained by deduction from general principles ^vhich are in turn derived from phenomena. It follows that Carnot is the founder not only of the second law of thermodynamics but also of thermodynamic deduction. In the former and fully recognized role Carnot is the first member of the trio which includes Clausius (1850) and William Thomson (1851); in the latter and at present unrecognized role he is the first member of a highly ramified hierarchy which culminates in Gibbs and among whose chief additional figures are: Clapeyron(i834), Helmholtz(i847, 1877!!.), William Thomson (1848 ff.), Clausius (1850), Kirchhoff (1858), Guldberg (1867 ff.), Massieu (1877), Boltzmann (1884), Van't Hoff (1884 ff.), Nernst (1888 ff.), Lewis (1907, 1923). That in brief is the significance of the forgotten theorems of Carnot for the history of science. This view of Carnot's role in the history of thermodynamic deduction might be criticized on the grounds that a majority of the forgotten theorems, al- though correct, were deduced by Carnot from the erroneous caloric theory (class B). The answer to this criticism is that the famous theorem on reversible engines was likewise so deduced by Carnot, and that therefore, as long as we regard Carnot as the founder of the second law of thermodynamics, we can hardly escape from regarding him also as the founder of thermodynamic deduction. The implications of this duality become clearer in the light of some familiar facts concerning the structure of theoretical physics. Any of the major sub- divisions of theoretical physics— for example, classical mechanics, relativity theorv, statistical mechanics, thermodynamics, etc.— has two parts: the first consists of the erection of general principles through expeditious discussion of selected phenomena; the second, of the systematic deductive exploitation of these general principles. This is often expressed by the statement that in theoretical physics we have induction followed by deduction. Actually, how- ever, the arguments leading to the general principles are more complex than mere induction in the strict sense: they seem to consist in general of expedient combinations of induction, deduction, and plausible assumption. Thus the argument leading to the second law of thermodynamics in the form Q^TdS contains all three of these elements, and in any system of statistical mechanics there is always at least one general principle, for instance, that of equal a priori probabilities, which is frankly an assuinption. We therefore prefer 28o Thermodynamic Theorems of Carnot to say that in the sciences making up theoretical physics we have construction followed by deduction. Turning now to consider the evolution of any of these sciences, we are not surprised to find that the early stages are given over largely to construction and that, as the science matures, deduction becomes increasingly important and eventually predominates. In some cases there has been reached a relatively final stage of high perfection, characterized by expositions from which constructive argument is entirely omitted and the general principles are treated solely as axioms sufficiently verified by their consequences; examples are the exposition of classical mechanics starting from Hamilton's principle and that of thermodynamics starting from the axioms of Carathcodory.* In such cases the constructive arguments have, however, not been banished completely, but have simply been relegated to the elementary textbooks. Within this general evolutionary pattern the position of Carnot is now clear: he stands at the beginning of thermodynamics with a constructive contribution of the first magnitude, supplemented by a deductive achieve- ment noteworthy for its prescience. IV. Significance for Thermodynamics at Present Needless to say, of the forgotten theorems of Carnot the true ones— those belonging to classes A and B— have not been forgotten in the sense that their content has become unfamiliar. What has been forgotten is the exist- ence of knowledge deducible from Carnot's theorem on reversible engines without the help of the first low of thermodynamics. Carnot's theorems of class A are examples of such knowledge and therefore have a special sig- nificance which is permanent, though neglected by contemporary authors. In order fully to reveal this significance we shall give below a summary of the consequences deducible from Carnot's theorem without benefit of the first law. The existence of such consequences was well known to the pioneers of thermodynamics, and most of the formulas in our summary can be found, more or less mingled with consequences of the caloric theory or of the first law of thermodynamics, in the writings of Clapeyron, Clausius, and above all William Thomson. We shall, however, be able to make only passing reference to the history of some of the individual formulas. The neglect of results of the type here summarized has probably been due to haste to obtain the more inclusive results that follow from the first and second laws combined, or, in the case of perfect gases, from the first law alone. But the need for such haste is now long past and a time has arrived in which writers on thermodynamics might profit from an aware- ness of the consequences that follow from Carnot's theorem without the first law. * See A. Lande: "Axiomatische Begriindung der Thermodynamik durch Cai atheodorv," in Handbuch der Physik, ed. by H. Geiger and K. Scheel, IX (Berlin: 1926), 281. F. O. Koenig 281 SUMMARY OF THE CONSEQUENCES OF CARNOT'S THEOREM ALONE 1. The Existence of Absolute Temperature. Let the different possible em- pirical temperature scales be numbered \, 2, . . . , i, . . . and let the values of a given temperature on these scales be denoted by 9^, 6.,, . . . , Oi, . . . respectively. Then Carnot's theorem can be expressed in the form ^ = 'i>,(e/,en (I) in which Ti^ is the work of a Carnot cycle whose higher and lower tem- peratures are 6i' and ^/' respectively, Q is the heat absorbed at the higher temperature, and i denotes a function of ^,/ and 6/' which is independent of the working substance. Since W/d is independent of the selection of temperature scales, it is furthermore clear that the value of $, for a given pair of temperatures is the same for all scales and that therefore the form , must vary from scale to scale, which is the reason for attaching the subscript i to the $. If we introduce the physically obvious assumption implicit in the work of Carnot and his successors, that as 6/— 6/' approaches zero, TT^ becomes an infinitesimal, SIF, of the same order as di'—Oi", eq. (1) yields, for a Carnot cycle working over the infinitesimal temperature interval dOi which includes the temperature 6i -^ = ^,{edde, (2) in which the function ixi{6i), named Carnot's function by Thomson, is independent of the working substance. Furthermore, while both the form of fxi and its value at a given temperature depend on the choice of the scale /, the value of the infinitesimal, ixi{6i)d6i, for a given temperature and a given infinitesimal temperatuie interval, is an invariant of all scales /. Hence the value of f J er for a given temperature interval is also an invariant of all scales /. Now con- sider the functions T^^i{6i) defined by the differential equations M{T)dT = ui,{di)Si (3) in which the form of M is arbitrary but the same for all scales /. Integra- tion gives F(r) - F{T") = ). o in • vo .s dC —- ^- - ^ • O Ifl » ■■- — (0 fEstrone I Mg- I mg. -16 D \Progesterone 2 H 3 fEstrone 10 Mg. I mg. — 21 D 1 Progesterone 2 H 3 Progesterone I mg. -25 D 2 H ID — II D 3 H 3 FSH 0.12 mg. -16 D A H 5 L';4iS 2 mg. — I DAB 4 H 3 Lj66B 4mg. 0 DAB 5 H 5 fL366B \Estrone 2mg. I Mg- + 2 LA +(4) D(i) 6 H 3 /L366B \FSH 2mg. 0.12 mg. -14 LA+ 7 H 3 L351B 5 mg. + 20 LA+ (2), D.^B (i) 7 H 3 LtciB 10 mg. +v LA+ 8 H 3 l'L35iB 1.2 mg. I Mg. +19 DAB \Estrone 8 H 3 /LiciB 2.5 mg. I Mg. +31 DAB \Estrone 8 H 6 /L151B. ... . 5mg. I Mg. + 21 LA+ + + (<) 1 Estrone • LA++ (i)" 9 H 3 /L-JciB 5 mg. 0.12 mg. + 10 LA+ + + (2) \FSH LA++ (i) ID HO 3 /L3S1B \ Estrone 5 mg. I Mg. + 11 DAB I I HO 3 /L151B \FSH 5 mg. 0.12 mg. + 24 DAB HO J fL-jciB 5 mg. 1 Mg. 2 mg. + 16 LA + + (i), 12 i Estrone LA+ (2) [Progesterone I? HO 3 'L384S ^ Estrone 2mg. I Mg. I mg. - 3 LA+ (2), DAB (i) (Progesterone * L.HiS. L366B, and L384S =purified lactogenic hormone; L.3;i=a crude lactogenic preparation, t D = regression, D.^B =maintenance, LA =lobulo-alveolar growth; number of rats in parentheses; plus signs indi- cate grade of LA development. 322 Mammary Growth After Hypophysectomy promoting factor. Rather, it has seemed important to determine the hormonal factors necessary to build as complete a mammary tree as is found in the ante- partum rat (LA-f++ or +-f++)- Experiments and Results (Table i) Experiment i. — Purified lactogenic hormone (2 mg. or 60 I.U. daily) was injected siibcu- taneously for 10 days into 3 normal female rats. The estrous cycles were inhibited, and at necropsy on day 11 the vaginae showed mucification, a single crop of large corpora lutea and small Graafian follicles were found in the ovaries, and the mammary glands showed complete lobulo-alveolar growth (pi. 1, fig. 5). Normal control rats sacrificed at 70 days of age showed a DAB condition in the mammary glands comparable to plate 1, figure 2. Experiment 2. — Estrone alone (10 I.U. and 100 I.U. daily), estrone (10 I.U. and 100 I.U.) plus 1 I.U. of progesterone daily, and progesterone alone (1.0 I.U. daily) were injected for 10 days into groups of 3 hvpophysectomized female rats. Except for the group injected with pro- gesterone alone, the rats showed estrous vaginae, but the mammary glands in every case showed regressive changes similar to those of hypophysectomized controls (pi. 1, fig. 1). All ol the rats lost ^veight. Experiment j. — A preparation of partially purified pituitary follicle-stimulating hormone (FSH) was injected subcutaneously into 3 rats at a daily dose of 5 R.U. for 10 days. This dose produced large follicles in the ovaries and kept the rats in estrus, but as with estrone the mam- mary glands sho^ved regression. The animals lost weight. Experiment 4. — Purified sheep lactogenic hormone ^vas injected at the 2 mg. (60 I.U.) level for 10 days into 5 hypophysectomized female rats and purified beef lactogenic hormone at the 4 mg. level (120 I.U.) for a similar period into 3 hypophysectomized rats. The mammary glands of all but one of these rats showed maintenance (DAB, pi. 1, fig. 2). One animal on the 2-mg. dose shoued regressing glands. On an average the rats maintained their preoperative ^veight in this experiment. Experiment 5. — Five hypophysectomized female rats were injected daily for 10 days with the same beef lactogenic hormone used in Experiment 4 (2 mg.) plus 10 I.U. estrone. Slight but definite lobulo-alveolar growth (LA-I-, pi. 1, fig. 3) was found in the mammary glands of all but one rat which showed regressive glands. This rat lost 10 gm., although the group showed a slight average gain. Experiment 6. — Three hypophysectomized female rats were injected daily for 10 days with 2 mg. of the same pine beef lactogenic hormone plus 5 R.U. of FSH. The mammary glands of all animals showed slight lobulo-alveolar giowth (LA-I-, pi. 1, fig. 3). There was an average weight loss of 14 gm. in this group. Experiment 7. — Two groups of 3 hypophysectomized female rats were injected daily for 10 days with 5 and 10 mg. respectively of a crude lactogenic preparation containing approxi- mately 10 I.U. of lactogenic hormone per mg., a Moon unit (normal immature male rat test) of adrenotrophin in about 60 mg. and a unit of .somatotrophin ("growth hormone") in about 1 mg. Slight lobulo-alveolar growth (LA-h) was observed in the mammary glands of five rats while the sixth animal (5-mg. dose) showed DAB. Experiment S. — Six hypoph)sectomized female rats were injected daily for 10 days ^vith 10 I.U. of estrone plus 5 mg. of crude lactogenic hormone. Excellent lobulo-alveolar growth (LA-H-f, fig. 5) was observed in 5 animals and good growth (LA-l-f, pi. 1, fig. 4) in the sixth. The rats showed an average gain of 2 gm. per day. Other experiments in ^vhich 1.2 and 2.5 mg. of the crude lactogenic extract were injected with 10 I.U. of estrone merely resulted in mam- mary gland maintenance (DAB) although the rats gained from 1 to 2 gm. per day in body weight. Experiment §>. — Three hypophysectomized female rats were injected daily for 10 days Avith 5 mg. of the same crude lactogenic preparation plus 5 R.U. of FSH. The results were similar to Wm. R. Lyons 023 those of Experiment 8, two of three rats showing maminary growth rated as LA— r-r and one showing LA++ growth. Experiment lo. — Three hypophysectoniized-oophorectomized rats Avere injected daily for lo days with lo I.U. of estrone plus 5 mg. of the crude lactogenic extract. No growth was in- duced in the mammary glands, but regression was prevented. The rals gained about 1 o-m. per day in weight. Experiment 11. — Three hypophysectomized-oophorectomized rats were injected daily for 10 days with 5 R.U. of FSH plus 5 mg. of crude lactogenic extract. The results were similar to those of Experiment 10. Experiment 12. — Three hypophysectomized-oophorectomized rats were injected daily for 10 days with 10 I.U. of estrone, 2 I.U. of progesterone, and 5 mg. of crude lactogenic hormone. The mammary grow th induced Avas rated as LA^^- in one animal (pi. 1, fig. 4) and LA+ in two. In another experiment in which 1 instead of 2 I.U. of progesterone was used, all three rats showed LA+ growth. The rats gained 1-2 gm. per day in body weight. Experiment 75. — Three hypophysectomized-oophorectomized rats Avere injected daily for 10 days with 10 I.U. of estrone, 1 I.U. of progesterone, and 60 I.U. of pure lactogenic hormone. The mammary growth induced was rated as barely positive lobulo-alveolar growth in two, while in the third rat, the gland was merely maintained. These rats lost 1-5 gm. in body weight. Discussion The doses of estrone and progesterone (alone or in combination) used in these experiments failed to stimulate any further growth of the hypophysectomized rats' mammary glands; and they were ineffective in even maintaining the slight alveolar development always present in these rats before hypophysectomy. This was also found to be true in similar animals treated with a pituitary FSH capable of stimtilating the secretion of estrin by the ovary. These findings are in accord with the results obtained by the majority of workers who have at- tempted to stimulate hypophysectomized rats' mammary glands \k\x1\ the sex steroids. Higher doses of the sex hormones in different ratios should be tried before concluding that they are completely ineffective in stimulating mam- mary growth in the hypophysectomized rat. It has been reported that these hormones do stimulate lobulo-alveolar mammary growth in the hypophysec- tomized rabbit.'" Purified lactogenic hormone injected into normal rats (2 mg. or 60 I.U. daily) caused excellent lobulo-alveolar mammary growth, but doses of 60-120 I.U. daily given to hypophysectomized rats were only capable of preventing or delaying postoperative mammary regression. It is known that lactogenic hormone stimulates the luteal cells to secrete progestin even in the hy- pophysectomized rat, but apparently in these animals the amount of estrin secreted is negligible or insufficient to act ^vith the progestin to cause mam- mary growth. This deduction is made from the fact that when 15-60 I.U of purified lactogenic hormone were given daily to hypophysectomized rats along with either 10 I.U. of estrone, or a pituitary FSH which caused estrin secre- tion, slight lobulo-alveolar growth was induced. The success of this type of treatment was undoubtedly partly due to the secretion of progestin by the activated corpora lutea, but it is also necessary to point out that hypophy- sectomized rats injected with lactogenic hormone consume more food than 324 Mammary Growth After Hypophysectomy uninjected hypophysectomized controls and maintain (on an average) their preoperative weight, in contrast to a loss of about 17 gm. by the latter. Never- theless, it was noted that minimal lobulo-alveolar growth was induced by lacto- genic hormone and FSH in some hypophysectomized rats which had lost as much weight as the controls and showed atrophic thyroids and adrenals at necropsy. In this same connection it has been noted that hypophysectomized rats injected with somatotrophic hormone may sometimes show bone growth even though they experience a loss of body weight. It was possible to obtain complete lobulo-alveolar mammary growth in hypophysectomized rats given 10 I.U. of estrone (or 5 R.U. of FSH) with 5 mg. of a crude acid-acetone beef anterior-lobe extract, L351B, containing approxi- mately 10 I.U. of lactogenic hormone per mg. Since a considerably higher unitage of lactogenic hormone given in the pure form with estrone was quite incapable of stimulating this degree of lobulo-alveolar development it would appear that another pituitary hormone (or hormones) important for mam- mary growth and not identified with the lactogenic is present in the crude preparation. It should be recalled that in Astwood's experiments,^ luteo- trophic and mammary-stimulating activities were not always found together in the same extracts. Enough somatotrophin was present in our crude extract to cause the hypophysectomized rats to grow at the rate of about 2 gm. per day even at a i-mg. daily dose, while the 5 mg. daily dose maintained the adrenals at about two-thirds of their normal weight. Further evidence for an unidentified pituitary factor or combination im- portant for mammary growth was found in the doubly operated rats. Estrone and progesterone caused slight lobulo-alveolar growth in hypophysectomized- oophorectomized rats when given with lactogenic hormone, and good lobulo- alveolar growth when given with the crude extract. The crude lactogenic preparation caused no lobulo-alveolar growth when given to the doubly operated animals with estrone or FSH. It is not possible to say, as yet, whether a new hormone or the proper combination of lactogenic, adrenotrophic, and somatotrophic hormones known to be present in the crude extract is capable of rendering the hypophysectomized rat's mammary gland responsive to es- trone and progesterone. A few experiments have already been carried out which show that no one of the three mentioned hormones injected with estrone will cause in the hypophysectomized rat the complete lobulo-alveolar growth induced with estrone and the crude combination. Apparently the mammogen theory of mammary growth is not supported by the data presented. A degree of lobulo-alveolar growth comparable to that considered adequate for assaying mammogen IF" was induced in (a) hypophy- sectomized rats injected with estrone and a pure protein* (the lactogenic hor- mone) and (b) hypophysectomized-oophorectomized rats injected with estrone, * The pure lactogenic hormone used was made by essentially the same method as reported in 1937.^^ We have shown that the hormone isolated by this method may be proven chemically pure by solubility tests.^*" The acid-acetone extract obtained before the NH3 step was used as the crude lactogenic preparation. Wm. R. Lyons 325 progesterone, and the same protein. In all our experiments the importance of the two ovarian hormones in stimulating prolactational mammary growth has been demonstrated. None of the well-identified pituitary hormones is at present suspected of being able to stimulate directly the growth of the mam- mary gland except the lactogenic. We have been able to stimulate "local" lactational growth in individual mammary glands (or sectors) in rabbits in- jected intraductally with pure lactogenic hormone. But it seems likely, judg- ing by the experience of most investigators, that the direct mammotrophic action of the lactogenic hormone is only possible in the hypophysectomized animal restored toward normalcy by certain of the depleted "target-organ" hormones, as for example, adrenocortical hormone. Conclusions I. In female rats hypophysectomized at 60-70 days of age and treated imme- diately thereafter, the following types of mammary development were ob- served after daily injections for ten days of the hormones listed: A. Regression to a gland showing bare ducts after: (1) 5-10 I.U. of estrone (2) 1 I.U. of progesterone (3) 10 I.U. of estrone plus 1-2 I.U. of progesterone (4) A pituitary FSH which stimulated the ovaries and produced estrous uteri and vaginae B. Maintenance of an apparently normal duct system bearing sparsely dis- tributed alveoli and alveolar buds after: (1) 60-120 I.U. of pvnified lactogenic hormone C. Incomplete lobulo-alveolar growth after: (1) 60 I.U. of purified lactogenic hormone plus 10 I.U. of estrone (2) 60 I.U. of purified lactogenic hormone plus 5 R.U. of FSH (3) 5 and 1 o mg. of a crude lactogenic hormone (also containing adieno- trophin and somatotrophin) D. Complete lobulo-alveolar growth after: (1) 5 mg. of a crude lactogenic extract also containing adrenotrophic and somatotrophic hormones, plus 10 I.U. of estrone (2) 5 mg. of the same crude lactogenic extract plus 5 R.U. of FSH II. In female rats hypophysectomized and oophorectomized at 60-70 days of age and treated immediately thereafter, the following types of mammary de- velopment were observed after daily injections for ten days of the hormones listed: A. Maintenance of an apparently normal duct system, bearing sparsely dis- tributed alveoli and alveolar buds after: (1) 5 mg. of the above-mentioned crude lactogenic extract plus lo I.U. of estrone (2) 5 mg. of the same crude lactogenic extract plus 5 R.U. of FSH 326 Mammary Growth After Hypophysectomy B. Incomplete lobulo-alveolar growth after: (1) 60 I.U. of purified lactogenic hormone plus 10 I.U. of estrone and 1-2 I.U. of progesterone (barely positive) (2) 5 mg. of the crude lactogenic extract plus 10 I.U. of estrone and 1-2 I.U. of progesterone (definitely positive) III. Normal 60-70-day-old female rats injected for ten days with 60 I.U. of purified lactogenic hormone daily showed complete lobulo-alveolar mammary development. ^ REFERENCES 1. Evans, H. M.: Harvey Lects. (1923) 19:212, 1925. 2. Evans, H. M., and Simpson, M. E.: Amer. Jl. Physiol. 89:51 1, 1931. 3. Lyons, W. R.; Simpson, M. E., and Evans, H. M.: Proc. Soc. Exper. Biol. & Med. 48:634, 1941. 4. Evans, H. M.; Simpson, M. E.; Lyons, W. R., and Turpeinin, K.: Endocrinology 28:933, 1941. 5. Nelson, W. O.: Anat. Recrd. 64(supp. 0:52, 1935. 6. Merckel, C, and Nelson, W. O.: Anat. Read. 76:391, 1940. 7. Lyons, W. R., and Pencharz, R. L: Proc. Soc. Exper. Biol. & Med. 33:589, 1936. 8. Folley, S. J.: Biol. Revs. 15:421, 1940. 9. Turner, C.W.: in Sex and Inlernal Secretions, ed. by E. Allen (2d ed.; Baltimore: 1939), 740. 10. Petersen, W. E.: Jl. Dair. Sci. 25:71, 1942. 11. Selye, H., and Collip, J. B.: Proc. Soc. Exper. Biol. & Med. 32: 1377, 1935. 12. Selye, H., and Collip, J. B.: Endocrinology 20:667, 1936. 13. Gomez, E. T., and Turner, C. W.: Proc. Soc. Exper. Biol. & Med. 37:607, 1938. 14. Lewis, A. A.; Gomez, E. T., and Turner, C. W.: Endocrinology 30:36, 1942. 15. Creep, R. O., and Stavely, H. E.: Endocrinology 29: 18, 1941. 16. Mixner, J. P., and Turner, C. W.: Endocrinology 29:324, 1941. 17. Lyons, W. R., and Sako, Y.: Proc. Soc. Exper. Biol. & Med. 44:398, 1940. 18. .\stwood, E. B.; Geschickter, C. F., and Rausch, E. O.: Amer. Jl. Anat. 61:373, 1937. 19. Samuels, L. T.; Reinecke, R. M., and Petersen, W. E.: Proc. Soc. Exper. Biol. & Med. 46: 379' '94'- 20. Nelson, W. O., and Tobin, C. E.: Anat. Recrd. 67(supp. i):iii, 1936. 21. Uyldert, I. E.; David, K. G., and Freud, J.: Act. brev. Neerland. physiol. 10:105, 1940. 22. Nathanson, I. T.: Shaw, D. T., and Franseen, C. C.: Proc. Soc. Exper. Biol. & Med. 42:652, '939- 23. Reece, R. P., and Leonard, S. L.: Endocrinology 29:297, 1941. 24. Gardner, W. U., and White, A.: Proc. Soc. Exper. Biol. & Med. 48:590, 1941. 25. Gardner, W. U.: Proc. Soc. Exper. Biol. & Med. 45:835, 1940. 26. Pencharz, R. L, and Lyons, W. R.: Proc. Soc. Exper. Biol. & Med. 38:388, 1938. 27. Asdell, S. A., and Seidenstein, H. R.: Proc. Soc. Exper. Biol. & Med. 32:931, 1935. 28. Astwood. E. B.: Endocrinology 28:309, 1941. 29. Lyons, W. R.: Proc. Soc. Exper. Biol. & Med. 35:645, 1937. 30. Li, C. H.; Lyons, W. R., and Evans, H. M.: Jl. Genrl. Physiol. 24:303, 1941. EXPLANATION OF PLATE PLATE 1 Figs. 1-6 represent one-half to two-thirds of the first right inguinal mammary gland removed from female rats after various treatments as listed below. Fixed in lo per cent formol, stained with alum-carmine, cleared in methyl salicylate, photogiaphed at x 1.25. Hypophysectomy and oophorectomy were performed on the same day when the rats \v'ere 60-70 days of age; treatment begun immediately and continued for 10 days with necropsy on day 11. Fig. 1. From a hypophysectomized rat injected with FSH, glucose, and thyroxin. This is the state of mammary development referred to in the text and table as D, signifying that little or no parenchyma beyond bare ducts is present. It is representative of the slow regression that occurs following hypophysectomy in the rat used, whether uninjected or injected with certain levels of estrin, progestin, or FSH. Fig. 2. From a hvpophysectoniized-oophorectomized rat injected with crude lactogenic hormone, FSH, thyroxin, and glucose. Referred to as DAB = ducts, and a few scattered alveoli which are mostly in the so-called "bud" form and not aggregated into lobules. The development shown here represents the upper limit of what may be found in normal 60-70-day-old rats. Fig. 3. From a hypophysectomized rat injected with purified lactogenic hor- mone, FSH, thyroxin, and glucose. Referred to as LA-I- = minimal lobulo- alveolar growth which may also be induced with pure lactogenic hormone and estrone. Fig. 4. From a hypophysectomized-oophorectomized rat injected with estrone, progesterone, crude lactogenic hormone, thyroxin, and glucose. Referred to as LA-H-, signifying an intermediate degree of lobulation. Fig. 5. From a normal rat injected with purified lactogenic hormone. Referred to as LA4-H-. This gland and that in figure 6 represent the variation seen in the pregnant rat from day 16 to term. The degree of development shown here has been attained in hypophysectomized rats injected with crude lactogenic hor- mone plus either FSH or estrone. Fig. 6. From a normal rat injected with crude lactogenic hormone. Referred to as LA I I I I = complete lobulo-alveolar growth. Enlargement of individual alveoli suggests that the early secretory phase had begun in this gland. [328] X .^>f y- :^^ i.}\ :^-^-v^.*.:;-v ^-^ ^ 4"^ [329] PULMONIC INTERSTITIAL EMPHYSEMA AND ITS SEQUELAE: AN ANATOMICAL INTERPRETATION By CHARLES C. MACKLIN AND MADGE T. MACKLIN X From the UNIVERSITY OF WESTERN ONTARIO MEDICAL SCHOOL LONDON, CANADA The authors regard it as a great honor to be invited to contribute to this testimonial to Professor Herbert McLean Evans who has done so much to advance functional anatomy, and to apply his discoveries to the solution of problems in scientific medicine. In this way he has materially aided the human race. We look back with pleastne to the days wiien we were all working to- gether in the great anatomical department of the Johns Hopkins Medical School under that master of anatomical research and development, the late Professor Franklin Paine Mall. We join in wishing for Professor Evans the continued possession of all factors requisite to the prolongation for many years of his outstanding work. TABLE OF CONTENTS i'Af;i. Introduction 337 Pulmonic Connective Tissue and Its Pneumatization ggy Human Pulmonic Interstitial Emphysema (PIE), Pneumomediastinum, etc 339 Movement of Air Bubbles Toward the Mediastinum 340 Airblock and Its Effects 340 The "Splinting" of the Lung by Incarcerated Air 342 Natural Relief of PIE 342 Mediastinal Emphysema and Its Alleviation 343 Animal Experiments 344 Forcible Overinflation of Part of a Living Animal's Lung Forcible Overinflation of Entire Fresh Lungs Analogies in Human Chest Pathology 346 Atelectasis Contributory Causes Survey of the Literature for PIE and Its Sequelae 348 Chest Conditions Probably Showing PIE and Sequelae, from Local Alveolar Overstrain 348 Atelectasis Neonatorum Autoresorption of Aberrant Air Why Aberrant Air Escapes Attention Interalveolar Pores Hereditary Predisposition Local Alveolar Overstrain PIE from General Alveolar Overstrain 352 Natural Overstrain Cases Abrupt Decompression Artificial Overstrain Cases Lung Blast Intratracheal Anesthesia and Artificial Respiration PIE Cases Miscellaneous Cases of PIE 358 Rupture Sites 359 Prevention, Diagnosis and Treatment 360 Summary 361 Final Note 365 References 366 [335] PULMONIC INTERSTITIAL EMPHYSEMA AND ITS SEQUELAE: AN ANATOMICAL INTERPRETATION * Iniroduction THE BETTER the Understanding of the functional anatomy of an organ the better the comprehension of its fundamental pathology. The newer knowl- edge, for instance, of the way the various parts of the lung work together to subserve its function has helped materially in the interpretation of its diseases, such as senile emphysema/ In a more restricted application of this principle, the functional conception of the connective-tissue formations of the lung has given us the means of giasping the nature and significance of pulmonic inter- stitial emphysema and its sequelae. In the development of this proposition we shall discuss the following among other points: how pulmonic interstitial em- physema (referred to hereinafter for brevity as PIE) originates and spreads, as revealed particularly by animal experimentation; some of the many clinical conditions with which it is associated; and the mode of action in producing its ill effects, known as "airblock," which may have serious and even fatal con- sequences. Although airblock is only now beginning to be diagnosed, its phys- ical basis, the aberrant air in the connective tissues of the lung, mediastinum and beyond, has been revealed at autopsy, and is now being demonstrated by roentgenography. From a study of pertinent clinical and pathological reports in the light of our laboratory experience we believe that airblock is present in serious degree more often than is realized, and hope that this em- phasis on its widespread occurrence will lead to its more general discovery and alleviation. Suggestions for its diagnosis and relief are given. First, let us say a word on the dynamics of pulmonic connective tissue. Pulmonic Connective Tissue and Its Pneumatization The subdivisions of the pulmonic connective tissue are named from their rela- tion to various lung structures; as subserous, inter- or perilobular (or septal), hilar, peribronchial, and perivascular. They are confluent, and the mass ex- tends directly, through the lung root, into the mediastinum and beyond. Viewed in toto, this tissue is a continuum of soft succulent material, readily dissectible by air which may gain entry into it. Most important of these sub- divisions in its bearing on PIE is the sheath of the pulmonary arterial and venous trees. This, in microsections, is seen as a layer between the vascular * In part as gi\cn by Charles C. Matklin on tlic in\ilati()ii pro^^^rain of the American Asso- ciation of Anatomists on April ii, igji, at the l'ni\ersity ol Chicago, inicler tlie title ■"Some features of the finictional anatoni) of the linig," and also befoie the Gibson Anatomical Society of the University of Buffalo Medical School, on March '51, 1942, on "Pulmonic Iiuer- stitial Emphysema, interpreted bv an Anatomist." Neither leduie has been published. n337 3 338 Pulmonic Interstitial Emphysema wall within and the ring of pulmonic alveoli without. Specifically, the outer boundary of this vascular sheath is made up of the contiguous bases of the perivascular alveoli which, collectively, form a mosaic or pavementlike layer. This extremely thin membrane is all that intervenes between the air of the lung and the fluid-filled connective tissue of the underlying vascular sheath. Now the interesting thing, functionally, about this perivascular ring of alveoli (or rather branched tunnel system, for the alveolar layer is continuous around the pulmonary blood vessels from the smaller branches to the larger trunks) is that it ividens with inspiration. This action has the effect of in- creasing the caliber of the pulmonary arteries and veins during inspiration, and thus of lowering the resistance of the pulmonary circulatory system. In other words, the inspiratory enlargement of this distensible perivascular en- velope of alveoli steepens the down-gradient from right ventricle to left atrivun. It is of less importance, at this juncture, to remark that this widening of the arterial and venous streambeds of the pulmonary system is accom- plished by the stroma pull of the lung, which is motivated by the pressure of the incoming air, which, in turn, is permitted to act thus because of the muscu- lar action of the inspiratory mechanism; that, through this functional anatom- ical conception, further light is shed on the fortunate natural arrangement whereby the operation which brings air into the lungs synchronously brings more blood into them; and that this explanation of an inspiratory "swelling action" on the pulmonary blood vessels makes less puzzling, perhaps, such cryptic terms as "negative pressure" and "suction." Rather for us, now, the important thing to note is that this perivascular tunnel system of alveoli does widen during inspiration with, consequentially, a certain amount of stretch- ing of each and every alveolar base which separates the air from the under- lying connective tissue of the vascular sheath; and that this stretching due to widening of the tunnel is augmented by that due to its lengthening, for inspiratory lengthening of the pulmonary blood vessels and their environ- mental structures is just as much a part of inflational adjustment as is the lengthening of the bronchial tree during the same respiratory phase.^ This much is physiology. As long as the swelling of the blood vessel keeps pace with the enlargement of the surrounding zone of alveoli there is no disturbance of the pressure rela- tions between the alveolar air and the underlying connective tissue; or, more explicitly, there is no tendency for air to break through the thin alveolar bases and invade the vascular sheath. All that happens in the sheath is a passive enlargement in diameter in accommodation with the similar enlargements of its neighbors. But if the alveolar zone becomes overdistended without there being a simultaneous and equivalent swelling of the blood vessel, then a down- ward pressure gradient is set up between air and connective tissue, and in con- sequence minute ruptures may be formed in the thin alveolar bases, weakened through divarication of their network of fibers, through which air enters the sheath. The same imbalance may occur if the blood vessel shrinks (as it does Macklin and Macklin ^gg when, for any reason, the normal volume of blood is not received from the right ventricle) without there being an equivalent narrowing of the alveolar envelope. These leakage-favoring factors are spoken of, respectively, as "A" and "B." When they act together there is even more tendency for air to leak into the vascular sheaths than when either is actino: alone. Much could be said as to the efficiency of the lung as an air container, and the safeguards against leakage of air into the pulmonic connective tissues usually obtaining, but this is not the place to say it. The air holder is not always impervious. We admit, too, that air may break out from the lung at other points than the perivascular alveolar bases, under abnormal conditions, as into the interlobular septa, or even through the pleura. We feel confident, how- ever, that the perivascular zone is an important one from the point of view of leakage of air and are convinced that a knowledge of the functional anatomy of these parts has done much to explain the mode of origin of PIE. From the \ascular sheaths the air can spread into the other parts of the lung connective tissue. It is through pressure upon the blood vessels by air bubbles, and through interference by air incarcerated in the connective tissues with the ventilational action of the lungs that PIE exerts its damaging influence on the vital func- tions of circtdation and respiration. Human Pulmonic Interstitial Emphysema, Pneumomediastinum^ Etc. This is just what happens in the human subject when air, in damaging quan- tity, enters the interstitial tissue of the lung. It is an "accident" of surprisingly frequent occurrence, and is found under a wide variety of clinical conditions. The initial step, pulmonary interstitial emphysema, is seldom recognized clinically, and not always at autopsy. Even emphysema of the mediastinum, the next point of invasion by the air, is only now beginning to be diagnosed clinically with any degree of frequency.""" The further extension of air into the subcutaneous tissues of the neck, face, chest, etc., has been recognized, of course, although the source of the air has not been well understood. It has usually been thought to be a laceration in the mucosa of the upper airway, in those cases in which intratracheal anesthesia has been used or in which some instrument has been introduced into the trachea. Unexplained entry of air into the pleural cavities has long been known as "spontaneous" or "idiopathic" pneumothorax, although here, again, the source and rotue of the air have not been correctly visualized in the majority of instances. Spontaneous pneumo- thorax has often been regarded as arising through rupture of subpleural blebs," the origin of which has been variously interpreted, rather than as oc- curring from the rupture of an air-distended mediastinum" which has received its air by extension from a PIE. Berkley and Coffen,'" however, recognized that subcutaneous emphysema and pneumothorax, arising in patients •with influenza, had a common source of origin, namely air in the mediastinum. Pneumoretroperitoneum has been looked upon as due to the extension of air from the mediastinum, although PIE as the origin of the mediastinal air has 340 Pulmonic Interstitial Emphysema often been overlooked. Pneumoprecordium has been diagnosed occasionally, but PIE as its source has not often been suspected. This condition is not in- frequently called pneumopericardium, we think, erroneously, for the most part. When the air has worked its way forward over the heart, a "crunching" sound, audible several feet away from the patient, and synchronous with the heart beat, may be heard.*' ^'^^ This is pathognomonic of air in the anterior medi- astinum, but absence of this sign does not rule out the presence of air in other parts of the mediastinum. Movement of Air Bubbles toward the Mediastinum Once rupture openings have been formed, the tendency is for leakage of air into the interstitial tissue to continue until conditions arise to prevent it. The air does not accumulate near the vascular-sheath invasion sites but tends to move toward the mediastinum, in which it may develop a pressure gieater than that of the atmosphere. The ordinary plus pressure of expiration woidd seem insufficient to account for the impulsion of the bubbles against such opposition, and elevation of the presstire head in the leaking alveoli, as by coughing and straining with the glottis closed, must be postulated in explana- tion. Newly entered bubbles push on those already in the connective tissue. Another factor in moving the bubbles along woidd seem to be a sort of "pumping" or "milking" action on the part of the broncho-vascular rays. One of the most interesting features of the functional anatomy of the lung is the way all structures radiating out from the hilum, such as bronchi and blood vessels, elongate in inspiration and shorten in expiration."'" Without this ac- commodational length change, inflation and deflation of the lung would be impossible.^' ""^ Yet this perpetual lengthening and shortening, particularly of the pulmonary arteries and veins, is not an unmixed blessing when once air has entered the sheaths, for it seems to have the effect of moving along the train of air bubbles toward the lung root until stopped by the mass of packed bubbles, and thus of augmenting the same propensity arising from their being pushed from behind by new bubbles leaking in under the raised pressure head. The stream does not seem to become easily dammed up; not, perhaps, at times, until the pressure in the mediastinum rises to a prohibitive degree, in cases where escape from that space is not readily provided. The path of least re- sistance is toward the root. The way is thus constantly open for more air to enter from the leakage points. The air bubbles increase in size through co- alescence and the degree of their pressure damage increases as they move along, for they constantly impinge on larger and larger trunks. Here, again, a knowl- edge of the functional anatomy of the lung aids in the understanding of PIE. AlRBLOCK AND ItS EFFECTS Airblock is primarily a malady of the chest, but since the aberrant air may extend through the superior and inferior thoracic apertures, a field beyond the mediastinum may develop in addition to the ptilmonic and mediastinal Macklin and Macklin 341 ones. Clinically the effect is general. Airblock strikes at two vital functions, circulation and respiration. More thought might well be given to the physical nature of the air bubbles which do the damage. Their stnface tension derives from the natuie of the connective-tissue fluid. The bubbles, as they emerge from the alveoli, and press upon the terminal pulmonic vascular branches, must be very tiny. Their tendency to meige is irresistible. In the mediastinum they are large and conspicuous, and in the retroperitoneum may be inches across. It may take some hours for any given case of airblock to develop. With conditions favorable for the continuance of air leakage the air tends to move on into the mediastinum. The air accumidates in the mediastinum and the cmious condition may arise in which its pressure there is even greater than that of the atmosphere. Situated thus it is in the position of a strong invading army in a vital center. It is placed where it can do the greatest possible damage. All the mediastinal contents are pressed upon. The filling of the ventricular, and particularly the atrial, chambers would be hampered, and, in extreme cases, actually prevented. The great vessels, too, would be pressed upon. Systemic venous return would be slowed and even halted. Generation of pressure in the mediastinum would reflect back on the lung, and the aberrant air pressure there would also rise. Relief would be instantly afforded by air breaking out of the mediastinum, so that, in some ways, such things as subctitaneous emphysema and retroperi- toneum are favorable signs in that they indicate at least a temporary surcease for the oppressed mediastinal contents. Pneumothorax is of the same order, and, in addition, if on the affected side, it tends to stop the leakage. The pro- gressively weakened circulation is less and less able to take care of the task of absorbing the air, so that there is little or no assuagement from that quarter. The importance of a serious case of pneumomediastinum can hardly be over- emphasized. It constitutes a dramatic crisis. Yet until recently many physicians have been blind to a realization of what was going on, and how the condition might be relieved. It has, no doubt, been occurring for ages, in the dark. It may be that there are many relatively slight cases of PIE w^hich clear up without producing serious airblock, but there are undoubtedly many cases where the interstitial emphysema rapidly advances from local to general as the bubbles, swarming toward the lung root, hamper the circulation in the entire lung rather than in a part; then, invading the mediastinum"""^ they proceed to evoke the circulatory crisis there, already described. Physicians are urged to visualize the effects on the mediastinal contents when the pressure of medi- astinal air rises above atmospheric pressure. Doubtless many of these cases were thought by the attending physicians to be hopeless, when but a needle prick would have saved their lives! Many were probably set down to a failing myo- cardium. In view of the hidden nature of the malady it is not surprising that there has been failure in diagnosis. Some have been mistaken for angina pec- toris,"'^' and the symptoms were presumably due to pressure upon the gieat vessels, heart, or perhaps specifically upon the coronaries themselves. Here is 342 Pulmonic Interstitial Emphysema a condition which, like syphilis, acting behind a mask, may simulate other diseases. Backing up from the mediastinum, a PIE may be established in the op- posite lung, and the distress of the patient become even more aggravated. The imagination should be called into play to envisage serious cases with pulmo- nary and systemic vessels impeded by the indentations of the air bubbles, and the action of the heart itself embarrassed or completely throttled by the mass of froth around it, as well as by the encroachment upon it of the bloated and stiffened lungs. Airblock should be regarded as a disease in itself. The "air sickness" of aviation medicine is not the only pneumapathy; here is one which is widespread and may be deadly. The condition may— indeed often does- clear up of itself, but this should not lead us to view it lightly. The hopeful side of the matter is that so many progressive physicians are studying PIE and mediastinal emphysema scientifically and arriving at useful means of diagnosis and treatment. The "Splinting" of the Lung by Incarcerated Air In ftmctional anatomy, the arrangement of the bronchial and vascular trees, with their connective-tissue envelopes, has been likened to a mechanical frame- work which opens up in inspiration and closes in expiration. The aforemen- tioned lengthening and shortening of these structures are an integral part of this action,"' without which no respiration could occur. For it to take place it is essential that there be complete freedom of inlet and outlet of air. When air is trapped in the vascular sheaths, interlobular septa, or other parts of the connective tissue, this opening and closing action is interfered with. The normal amount of air intake is reduced, the requisite collapse is prevented, and the ventilational process hampered; in short, the lung is more or less seriously fixed or locked in a state of bulk comparable to partial or full inspiration.-^ The enlargement of its mass is, in man, more notable in the cen- tral and radicular than in the peripheral regions. The greater the accumula- tion of air, the more the patient strives by forced breathing, to overcome the handicap, but too often succeeds only in making the condition worse. So a vicious circle is set up. In extreme cases the chest is fixed in hyperinspiration, and the tidal flow of air is practically nil. Much of the baneful influence of PIE is due to the immobilization of the lung from these air bubbles, which interdict the normal operation of the aforementioned pulmonic framework or "skeleton." Once again functional anatomy has assisted in the interpreta- tion of a pathological condition. This air locking or splinting effect on the lung is an important fraction of the basis of airblock. Natural Relief of PIE In the ordinary course of events there are but two means of mitigating PIE— by absorption into the blood stream, and by mass escape. It does not seem that diffusion into the airway plays any part. With the circulatory embarrassment inevitably caused by PIE there is diminished power of absorbing aberrant Macklin and Macklin 343 air. The mass escape may be through a rupture in the visceral plema, but we have found no evidence of this mode except under drastic experimental con- ditions not comparable to clinical ones. In otir experience it is via the hmg root into the mediastinum, so widening the airblock field. It is certain that air bubbles are forced into the mediastinum, traveling thence along the sheaths of the great pulmonary blood vessels. Unfortunately, there is a considerable amount of air which cannot be so evacuated, because, as we have just seen, it is trapped in regions of the connective tissue of the lung such as the inter- lobular septa. If the pulmonic airblock involves one lung only, its prognosis is less serious than when both lungs are implicated. No artifice has been sug- gested for removing the air from the lung interstitium. Mediastinal Emphysema and Its Alleviation Mediastinal emphysema, or pneumomediastinum as it is often called, is seen to be the result of invasion of air, via the lung root, from an initial PIE. The tendency is for alveolar-base air leakage to go on, once it gets started, until a counterbalancing condition obtains. Air-bubble columns continue to move along lines of least resistance toward the lung root, and so a mediastinal cm- ])hysema is likely to grow in severity. The pressure rises, and cardiac and vascu- lar functions are handicapped.'' Easement may be naturally afforded to the compressed contents by efflux of air through perivascular channels which it dissects for itself into the root of the neck and retroperitoneum. The holding strength of the mediastinal walls may be taxed to the breaking point. In rab- bits a rupture of the lateral mediastinal wall, giving rise to a pneumothorax, has been demonstrated.'" Instant and dramatic deliverance of the stressed medi- astinal contents is effected by such a sudden breakout of air. A pneumothorax, too, on the side of an increasing PIE, has the effect of stopping the leakage, and hence artificial pneumothorax, at least partial, has been suggested as a therapeutic measure.""'* Lateral roentgenograms are useful in revealing pre- cordial air pockets, which may be evacuated by means of a hollow needle.'' The anterior mediastinum is the place of election for direct attack upon the cause of airblock. An understanding of the morphodynamics of airblock has led to suggestions,""-* already successful,'' ' for its alleviation. It seems probable that lives have already been saved by recourse to intervention. Again, func- tional anatomy has aided pathology and also practical medicine. Hyperatmospheric Mediastinal Pressure.— Olx\\n?^x\\y the pressure in the mediastinum is lower than that of the atmosphere.'" In some cases of pneu- momediastinum it becomes higher. For instance, in Gumbiner and Cutler's case' the plunger of the syringe shot out when they tapped the substernal pocket of air, and this w\as in a newborn babe. How can we account for this dramatic elevation? The mechanism seems analogous to that in the so-called "pressure pneumothorax." The motive power, in nonexpcrimcntal cases, must ultimately be in the individual's own respiratory musculature, regular and accessory. Normally the interstitial tissue of the lung and mediastinum is pro- 344 Pulmonic Interstitial Emphysema tected against air invasion, the alveolar bases being competent to hold the air in the alveoli against not only the normal hyperatmospheric pressures of ex- piration, but the still higher ones of coughing, straining, forcible expiration, etc. But we have seen that when these bases are ruptured the tendency is for air to continue to leak into the interstitial tissue and for pressure to rise in the accumidated air pools. Once in the interstitial tissue the air apparently cannot go back into the alveoli. There is a sort of valve action here, and bubbles move toward the mediastinum in natural course. The augmented pressure head in the alveoli during covighing, straining, etc., is carried over into the bubbles and, in part, held therein. The effect is cumulative. It extends along the train of bubbles into the mediastinum. The potential raised in the elastic tissue of the broncho-vascular rays and other parts of the lung in inspiration, particu- larly when forced, is expended in the expiratory phase to pack the bubbles and move them onward, as we have seen in the discussion of the mechanism of transpulmonic air-bubble movement. It seems that air can in this way get into the mediastinum more easily than it can break out. It is thus compressed by the recoil of the chest, particularly after the inspiratory movements are forced, as in dyspnea. We have here something analogous to the intake and compression strokes of the internal combustion engine. The greater the air- block the greater the dyspnea; and the severer the dyspnea becomes, the more likely is the tension in the mediastinum to increase. It is to be noted that, as with tension pneumothorax, aberrant air can enter the mediastinum only during the expiratory phase, once the pressure in it has reached that of the atmosphere, for only at that period is a pressure in excess of atmospheric gen- erated in the lung; and not (as most authors erroneously state for tension pneumothorax) during inspiration. Therefore, when a patient, in whom atelectasis may be present, is making or has just made forced expiratory efforts of any kind, as in coughing, the clinician should be alive to the ever-present menace of PIE and particularly of its sequel— air in the mediastinum under pressure— which will certainly establish an airblock. If the leakage continues imder such conditions, ruptiue from the mediastinum into the pleural cavity is likely to set up a tension pneumothorax. Animal Experiments The interpretation of PIE and its sequels just given has a broad foundation of data from animal experiments. Without the information gained from these we should not understand how air breaks out from the lungs; is moved in a definite way to distant points, and exerts its pernicious, even lethal, effects; nor should we be able to visualize the tissue mechanics of inspiratory swelling of the pulmonic blood vessels. Functional anatomy and, through it, pathology and clinical medicine are incalculably indebted to the experimental method. Through it the connective tissue of the lung is seen in a new light. As has so often happened in research work, the experiments were started with quite another objective— information on the nature of the pulmonic alveolar walls— Macklin and Macklin 245 and the discovery of the origin and ramifications of PIE was at first incidental; but the leads were quickly followed up, and what was initially a by-product became one of importance. There were two series: (a) local and (b) general overinflation of the lungs. Forcible Overinflation of Part of a Living Animal's Lung.—Some thirty-five animals were used, mostly cats. A flexible male urethral catheter, with the end cut off, was passed down the trachea of the anesthetized animal and as far as it would go without effort into the bronchus of the right lower lobe, which it entered naturally. The lung tissue about the tip of the catheter was then in- sufflated with pure air blown into it from the mains of the building. In the first cats used, the air contained the fumes of osmium tetroxide, but this made no essential difiierence. The pressure varied from i to 22 mm. of mercury as read on a manometer, but must have been somewhat higher than that to which the alveoli were subjected. In most cases the abdomen and thorax became swollen, and also the neck, axillae and chest if the insufflation was continued. The efi^ect was very striking. It was usual to find a double pneumothorax. Large air blebs were conspicuous in the axillae and retroperitoneum.^ The heart was surrounded by bubbles. In the fresh condition bubbles could be seen in the lung roots, and, by probing from the heart it was found that these were in the sheaths of the pulmonary blood vessels. Examination of gross slices and microscopic sections of the lungs, fixed by intratracheal injection, disclosed air bubbles along the sheaths of the pulmonic blood vessels and in the tissue extending out from these. The bubbles were small at the periphery of the lung, where the vessels were small, and increased in size toward the hilum. The appearance of the pulmonic vessels was indeed striking. The walls were col- lapsed, and might even be folded inward upon themselves until in many cases there was no discernible limien. Between this collapsed tube and the surround- ing lung alveoli was a large space occupied by air (or by a space which had formerly contained air when this had been given up in the technical manipula- tions), and this space was crisscrossed by delicate cobweblike strands, the remains of the connective-tissue sheath of the vessel. Looked at lengthwise in gross pieces, with a low-power lens, the vessels often appeared as mere crumpled tubes swung by tenuous guy ropes attaching in all directions, and surrounded by bubbles. A glance would convince anyone that little or no circulation could go on in such vessels. In one such case from an overinflated excised calf lung, fluid could not be forced through at a pressure of three feet of water. One could readily see why cyanosis should be a prominent feature of airblock. A similar occlusion of pulmonic blood vessels was found in the lungs of two children, one of whom had died after aspirating peanut frag- ments,^' and the other of atelectasis of the newborn.^" The alveolar-base rupture points are invisible in ordinary histological sec- tions, made from immersion-fixed blocks, but in certain of these animals their sites were located by a method of injection of the previously overinflated lobe, through the same bronchus, with hot gelatin containing minute carmine par- 346 Pulmonic Interstitial Emphysema ticles.'^ '* The alveolar bases about the finer pulmonary blood vessels of the •overinflated region were implicated. The role of the pulmonic connective tissue as an air conduit to the mediastinum was clearly demonstrated. Conditions here are not, of course, identical with those in the human sub- ject. The air pressure was sometimes much higher, for instance, than that under natural conditions, and the inflow was sudden and continuous. The overdistended region, however, was comparable to a region of compensatory emphysema such as frequently occurs in human lungs and where leakage of air into the interstitium is known to happen at times.^' Forcible Overinflation of Entire Fresh Limgs.—The fresh excised lungs of calves were overinflated n? toto"'' by blowing air into the tightly cannulated trachea until the alveoli ruptured. As the pletira became taut and the air escaped into the connective tissue and opened up runways therein, a striking noise of coarse crepitation was heard. Soon air began to bubble out from around the stumps of the great pulmonary vessels in the lung roots. It was a dramatic sight to watch it emerging thus continuously as long as it was forced into the trachea, and was an unforgettable lesson in the way air invades the mediastinum from the lung tissue. When the trachea was again opened, part of the air escaped through it, but the lungs did not resume their original size, and their enlargement was due to retained air in the interstitial tissues. In the lungs of cattle the partitions between lobule groups are very obvious, and these became infiltrated with air, giving a characteristic pavement appearance not only to the surface but to sections of the interior cut from intrabronchially fixed material. It was easy to see why a patient in whom such a condition had arisen should show a chest fixed in a state of inspiration, with little tidal flow of air, and also why dyspnea and cyanosis shoidd be prominent features of the airblock syndrome. Splinting of the lung and clamping down upon the blood vessels combine to vitiate the patient's circulatory and respiratory systems. In this material a few actual ruptures in alveolar bases not only of the vascular sheaths but of the interlobular septa were demonstrated in microsections. Analogies in Human Chest Pathology The lessons learned from the animal experiments go far to explain how air breaks into the pulmonic connective tissue, moves toward the mediastinum, and exerts its evil influence in airblock. A number of common factors underlie the animal and human phenomena. The human cases may be classified broadly in accordance with the two experimental subdivisions, namely, as local and general overstrain of the air-container system. Thus there is a large group which shows the result of local overstrain aris- ing as a consequence of definite pathological change. There is a limit beyond which the pleural cavity may not normally be reduced in capacity, and the lung has to adapt itself to this minimal volume. Hence, if any part of the lung is abnormally reduced in bulk, as it is in atelectasis, then other parts have to overexpand to fill up the space. This overfilling of a part with air is known Macklin and Macklin qa^ technically as "compensatory emphysema."* In this condition the individual air spaces are overdistended; and this means that the bases of alveoli lying against connective tissue, particularly that of the sheaths of the blood vessels, as we have seen, are strained and may become ruptured, allowing air to leak into the interstitium. Factor "A," with its downgradient from alveolus to vascular sheath, is then operative. Atelectasis— It is, then, quite possible for PIE to originate in any local, or alveolar, emphysematous area; and this brings strongly to the fore the im- portance of atelectasis as an initiatory cause of PIE. Atelectasis, as is well known, is found under a variety of conditions, but obstruction of the bronchus supplying the atelectatic region is a generally recognized immediate cause. An obstruction can arise suddenly, as by impaction of a foreign body in the bronchus; or gradually, in the course of disease, as in bronchopneumonia, when it results from inflammatory tumefaction and exudate. The last type is usually multiple in locale. Thus, atelectasis may be in one or many regions of the lung: it may involve a large area of lung substance or a small one. We feel that, if all human PIE cases could be searched through to their origin, it would be found that atelectasis, by inducing compensatory alveolar emphy- sema, is at the bottom of many, if not most, of them. Contrihiitory Causes— Now it is, perhaps, too much to say that all cases of local overstrain of the air container are followed by PIE, even in minute degree, and it would seem that something more than simple alveolar ectasia is usually necessary to induce a leakage of air into the interstitium. It is prob- able that the strain upon the alveolar bases is severe when the extent of the stretching is great, and particularly when it is accompanied by elevation of pressure, as in coughing and intense muscular effort. Where there is inflam- mation the alveolar bases may be presumed to be weakened from this state, and so may be more easily ruptured. In some individuals the bases are ap- parently naturally weaker, and so confer a predisposition to what may be termed "idiopathic" PIE. Then, too, Factor "B" must be kept in mind. It operates when, for any reason, the volume of blood in the pulmonary vessels is reduced, for then the internal buttress of the vascular sheath gives way, creating a further downward gradient from alveolus to sheath. It seems quite likely that if these vessels are overswollen the tendency to PIE resulting from alveolar ectasia may be warded off. Such a protection may be operating when, for instance, a lobe, or even one entire lung, is removed, for then the blood vessels of the remaining parts are carrying not only their own quota of blood but that from the parts which have been removed, and are thus engorged. Augmented in size in this way they encroach upon the space of their surround- ing sheaths and so tend to make up for the out^vard pull of the surroimding * It is unfortunate that pathologists and chnicians use the same term, "emphysema," for two different entities: (i) alveolar ectasia and (2) interstitial emphysema. Emphysema means to blow up or inflate. By using qualifying adjectives and speaking of "alveolar" emphysema on the one hand and "interstitial" emphysema on the other, \\c can do nuuh to avoid confusion. 348 Pulmonic Interstitial Emphysema ring of alveoli, which arises inevitably with the compensatory alveolar emphy- sema occurring by reason of the fact that the surviving lung substance has to enlarge to fill the space vacated by the ablated part. PIE has not so far been presented as an important postoperative problem in lobectomies and pneu- monectomies. So, pulmonic local overstrain, arising in an atelectasis-engen- dered alveolar ectasia, when augmented by contributory factors, conditions PIE. In this light, atelectasis is a menace. All possible means should be taken to diagnose it and be on the watch for its offspring, PIE, which makes itself known clinically by the airblock syndrome, and in other ways. Survey of the Literature for PIE and Its Sequelae A survey of pathological and clinical literature shows that there is reason to conclude that PIE was present in many cases where, to the attending physi- cians, it was occult; and, indeed, to believe that it is much more prevalent than is supposed. The evidence is direct and indirect. The direct, or pathological, evidence is obtained at autopsy when bubbles of air are disclosed in the inter- stitial tissues of the lung. The indirect, or clinical, evidence is of several sorts. Important is the roentgenographic display of the effect of air in the medi- astinum, which must have originated from a PIE; and also of the result of air in the pleural cavity, for this may ultimately have come from a PIE. Actual proof that there was air in the anterior mediastinum would be afforded by withdrawing it from the location with a hollow needle. The presence of sub- cutaneous emphysema would probably implicate the mediastinum as the source, and ultimately the lung interstitium. The symptoms and signs point- ing to the existence of airblock would be important indicators of PIE and its sequelae. The following resume of published clinical reports was prepared by one of us (M.T.M.) We shall first take up cases in which it seems reasonable to believe that local overstrain was the important causal factor, and where the air-pervious region probably arose as a consequence of an atelectatic condition. The survey makes no pretension to completeness. It is impossible here to men- tion all the clinical instances of this kind in the literature, and the references given will be only a few of the many that might be cited. We feel that this study brings home strongly the great useftdness of re- porting clinical and pathological observations, even though their signifi- cance is not immediately apparent. A number of interesting reports came from American army camps during the first World War when those who made them were doubtless hard pressed to find the necessary time for this work. These writers are to be warmly commended for their enterprise. Chest Conditions Probably Showing PIE and Sequels from Local Alveolar Overstrain Influenza is an outstanding example of diseases in this class. Torrey and Grosh^ reported patients in the epidemic of 1918-1919 who had dyspnea, cyanosis, fixation of the chest in a position of maximal inspiration, with Macklin and Macklin ^^q scarcely any tidal flow of air, and retrosternal pain. When these patients died, as was usually the case, there was not enough involvement of the lung to ex- plain the marked respiratory distress. There were areas of consolidation, with other areas of compensatory hyperinflation.^" The lungs did not collapse as readily or as fully as did normal lungs when the thorax was opened, and there were often numerous blebs, usually unruptured, beneath the pleura.'" Some- times these moribund patients showed sudden improvement, the cyanosis less- ened, the breathing became much easier and of fuller excursion, and the patient usually went on to recovery. The condition which always accompanied, and seemed to explain, this sudden improvement was the appearance of sub- cutaneous emphysema of the neck and face. The implications of this were not realized, and no attempt was made to remove the air from the mediastinum, thus relieving the pressure which was causing the symptoms. It seems to us likely, however, that had such a procedure been followed, many of those dying could have been saved. These various authors reported over 1,100 such patients, thus demonstrating that this condition is not uncommon. A very interesting finding was that of "air streaks" along the course of the pulmonary blood vessels in roentgenograms taken of influenza patients.^ This, in our view, is the first X-ray demonstration of perivascular PIE. Air bubbles along the vessels of the hand were reported as having been seen in such pa- tients.^' Again, the importance of these findings was not generally recognized. The simultaneous appearance of pneumothorax and subcutaneous emphysema in some of the patients was interpreted by Berkley and Coffen" as being de- pendent upon a single cause, namely, air in the mediastinum which had reached that region from the pulmonic "air streaks" already mentioned. We feel that these authors should be credited not only with the first X-ray demon- stration of PIE but with the discovery that PIE causes mediastinal emphysema and, through this, subcutaneous emphysema and pneumothorax. This impor- tant research came out of an American army camp during the first World War. Apart from epidemic influenza, there are reports of other infectious diseases involving bronchopneumonitis with probable atelectasis which provide evi- dence of some variety that PIE was present. Among such may be mentioned measles,''* pneumonia,'^'' bronchopneumonia*" including the postoperative type, and diphtheria." *" This great group of bronchopneumonic inflammatory con- ditions is most important, and careful watch for symptoms and signs of air- block should be kept so that immediate steps may be taken to relieve it. There are also conditions which, though not infectious in themselves, yet nevertheless lead to atelectasis through the mechanical disturbance they oc- casion or through inflammatory reactions they set up in the bronchi; and so, because of the area or areas of compensatory emphysema which ensue, they result in PIE and its sequelae. In this category are the foreign body impac- tions,"" occlusion of a bronchus by tuberculous lymph nodes,"^ or possibly cancer, scarring of the lung tissue by silicosis,'" or tuberculosis.*' Particularly in the cases of atelectasis which occur suddenly is there danger of leakage. 350 Pulmonic Interstitial Emphysema Atelectasis Neo7uitonim.—Ate\ectasis of the newborn is a special condition which has recently come in for some intelligent investigation and operative interference." ■'^ ^^ It is evident that airblock is quite prevalent in this class. One might well speak of "PIE neonatorum." Certain progressive physicians who have become aware of the common occurrence and serious nature of this condition have examined, particularly by X ray, newborn infants who were showing signs of respiratory distress, for the presence of air in the medias- tinum. The lateral roentgenograms have been particularly useful in revealing the signs of air in the precordial region, from whence it was withdrawn with a hollow needle.' The dyspnea and cyanosis have in this way been strikingly relieved, and children who have appeared moribund have recovered.' ' Not only has the air in the mediastinum been thus withdrawn, but it would appear from roentgenograms^ that air that has been in the connective tissues of the lung, particularly in the vascular sheaths splinting it and holding it in a position of inspiration, has to some extent been allowed to escape into the mediastinum, whence it, also, could be withdrawn. This method of treatment would seem best adapted for immediate and substantial relief of airblock, draining off some of the air even from the interstitial tissues of the lungs, and so relieving the disastrous effects of pressure on the intrapulmonary blood vessels as well as on mediastinal structures. Doubtless also some air escapes in time from pulmonic connective tissue apart from that of the vascular sheaths, such as the interlobular septa. Autoresorption of Aberrant Air.— It has been averred by a nimiber of work- ers*'^ that pneumomediastinum (and inferentially PIE) is a benign condition, requiring no special treatment, as the air is resorbed in a few days or at most weeks. This is doubtless true in some of the cases, but sometimes the patient does not live long enough to resorb the air, but dies of asphyxia before his unaided efforts can accomplish this desideratum. In the presence of such tragedies it seems more than possible that many physicians have not realized what was the real cause of death. Assistance in the form of prompt and ade- quate withdrawal of the air from the anterior mediastinum may tide the patient over an illness not necessarily in itself fatal in that patient at that par- ticular time, but in the course of which he may succumb because, unaided by surgical intervention, his circulation and respiration are too markedly im- peded by the aberrant air. Why Aberrant Air Escapes Attention.— It might be asked at this point why PIE and its sequelae have not been more often demonstrated in persons dying of airblock. There are several reasons why it may have been overlooked, among which may be mentioned the following: (i) autopsies, in proportion to the total number dying, are not often done; (2) the autopsy is usually long de- layed, so that a considerable amount of the air has diffused away after death; (3) the air bubbles are obscured by blood and serum; (4) the operation of cut- ting the lung bursts the bubbles and dissipates the air; (5) the pathologist was not actually seeking air in the pulmonic vascular sheaths and other parts of Macklin and Macklin 351 the lung connective tissue, and so o\erlookecl what was not to him obvious; (6) the lung was not fixed by intratracheal injection, but by immersion of small blocks of tissue, and the air was squeezed out of these or diffused away in the various technical manipulations. We feel that intratracheal fixation is very important in the demonstration of air bubbles and the impressions made by these in the pulmonic connective tissue. The lung, however, should not be overdistended with fixative, for then the vascular sheaths arc dilated, giving rise to an appearance simulating air in the sheaths.' Fisher's'" case, autopsied within an hour or so after death, revealed the air in the mediastinum and along the vessels very clearly. The case of Fisher and Macklin^' is another good example of the value of prompt postmortem examination and intrabronchial fixation. Interalveolar Po/rj.— Recent investigations in functional anatomy have shown that interalveolar pores are an actuality.'""'' It has been suggested that they allow air to diffuse into a region of lung whose bronchus has become occluded and that thus they tend to prevent atelectasis. It may be that they have some influence of this kind, but, if so, it is undeniable that atelectasis occurs in spite of them. This may be, in small areas, because the pores have become stopped up from the effect of inflammation; or in large areas because the pores do not occur in the connective-tissue septa lying between secondary lobules of lung substance. It is possible, too, that, when atelectasis does occur, the consequent compensatory emphysema involves less alveolar overstrain since, due to the free diffusion of air through the lung tissue on account of the jiores, the area of alveolar ectasia is spread out more and so the alveoli in any local area are not subjected to a breaking strain. Be that as it may, it is possible that the presence of pores explains the freedom of some individuals from PIE and its sequels when showing the leakage-predisposing conditions mentioned in the paragraphs immediately preceding. We would certainly not say that every person who develops these conditions will as a consequence have a PIE. There is, however, no reason to think that pores are more abundant in the lungs of some people than in others of the same age, so there seems to be no sound explanation for immunity to PIE, if so it may be called, on the part of a certain class of patient, on the basis purely of pore equipment. Hereditary Predisposition— On the other hand, it may be that susceptibility to air leakage may rest, to some extent at least, on hereditary ground. This applies not only to the cases already discussed, but to those which are to follow. There may be an inherent weakness in the alveolar walls of some persons, leading them to rupture more easily than normal, as evidenced by the numer- ous records of recurrent pneumothorax'' and of familial pneumothorax.'^ Local Alveolar Overstrain.— Yrova these reports and many others that could be cited, it appears reasonable to conclude that PIE is a concomitant of many disease conditions and bronchial accidents that involve local alveolar over- strain. We may affirm that any condition xvhich induces atelectasis inay pre- cipilate compensatory overinflaiion, and this, in turn, particularly when aided 352 Pulmonic Interstitial Emphysema by accessory causes, may lead to PIE arid its sequelae. The accessory causes already outlined are regarded as important. They may, indeed, be the deter- mining factors in the precipitation of any given case of PIE, in the presence of alveolar-base overstrain. Thus cough, straining at stool, etc., by raising the intra-alveolar pressure in a region of alveolar ectasia, particularly if the bases of the alveoli are weakened by an inflammatory process, may produce a rupturing of the alveolar bases and leakage of air into the interstitium. In- creased pressure in the alveoli is of great moment in this connection. Can it alone cause PIE? That seems possible, as will appear from the discussion of later cases. Certainly the combination of increased alveolar pressure and over- distention of alveoli is a dangerous one. Again it is brought home that the functional anatomy of the lung is neces- sary to an understanding of the response of the parts of the organ to abnormal conditions. Aided by the data from experimental work we are enabled to understand the origin and course of PIE, and even to predict in what type of disease condition PIE may be expected. The testimony of recorded clinical experience sustains our predictions. PIE FROM General Alveolar Overstrain So far we have dealt with clinical reports of human cases in which there may be presumed to have been a PIE resulting from some local overstrain of the alveolar bases, with complications. Even in these there may have been some general overstrain as well as the predominating local one. In them elevation of alveolar pressure, even though momentary, was regarded as very important. We now wish to consider cases in which it is reasonable to assume that there has been a general overstrain of alveolar bases on account of elevated pressure. It is convenient to divide these into two categories: (i) Natural, and (2) Artificial. Natural Overstrain Cases.~^he natural overstrain cases all show a severe muscular effort characteristically for a short period, but one which may be repeated; and when presumably the glottis is closed, so that the intrathoracic pressure is raised and blood is prevented from entering the thorax, so that the blood volume of the pulmonary vascular system is lowered and Factor "B" is produced. Manifestations of PIE have been found in individuals who have held their breath while lifting heavy weights," or tugging at heavy objects.^ If strong expiratory efforts are made while the breath is being held the intra- alveolar pressure is augmented; we have evidence of PIE occurring in women during childbirth,^"' ^^ or in persons straining violently at stool.^^ Closely an- alogous are the cases which occur even if the glottis is not completely closed yet where there is sufficient obstruction in the airway to raise the intra-alveolar pressure as expiratory efforts are made; then this pressure, combined with the reduction in caliber of the pulmonary blood vessels arising through the pro- longed interruption of venous return to the heart, will cause PIE. Thus we find that PIE has been reported in patients with asthma;^^ with congenital Macklin and Macklin 353 stenosis of the larynx, or in obstruction to the outflow of air from the bronchial tree of any variety whatever;'^ with violent cough whatever its origin;"' '■•*^ and also in patients who have been blowing against an obstruction, as into a wind instrument or a blow gun.'* In all of these, elevated intra-alvcolar pres- sure associated with Factor "B" may be postulated. The possibility of atelec- tasis with compensatory emphysema exists in some of these cases with lung disease, as indeed it does in any of these cases; biu it seems reasonable to assume that such simple conditions as weightlifting, childbirth, etc., would be unassociated with it as a general rule. The cases of prolonged muscular exertion, but without the holding of the breath or any obvious elevation of the intra-alveolar pressure, should be men- tioned here. Scott"' found pneumomediastinum, or pneumothorax, in athletes who had run a long hard race, some hours after the race was over, and here, we feel, PIE must have been the precursor. Failing heart, with supervening Factor "B," can hardly be ruled out here. The lungs had been ventilated to the full for a long period. There was no suggestion of atelectasis, though we think this a possibility. Abrupt Decompression.— On the borderline between the natural and arti- ficial overstrain cases are those which show PIE from too abrupt a decompres- sion, as in leaving caissons'^ or escaping from submarines. The latter is of particular naval importance at the moment of writing, and is an accident which occurs in men being trained in the technique of ascending from deep levels of water, when they fail to obey orders.™"' The recruits are taught to re- lease themselves from diving bells at various levels, and to ascend with relative slowness to the surface, pausing at different depths, and all the while to breathe in and out of a bag attached to the mouth. The nose is held by a clip, and the bag has a flutter-valve attachment which permits adjustment of pressure wathin the bag and thus within the lungs as the person ascends. This prevents the overexpansion of the air in the alveoli as the pressure on the outside of the chest lessens during the ascent, as would be the case if the man held his breath. Now this impulsion to hold the breath during the ascent is said to be over- mastering in some men, and when this is done during rapid ascent the air in the lungs expands, causing generalized increase in pressure and such over- inflation as is possible in the confines of the thorax. At the same time the pul- monary blood-vessel caliber is reduced on account of the obstruction to inlet of blood into the thorax. Factor "B" would thus be in operation, as w^ell as Factor "A" to some extent. In these cases, not only does air make its way into the interstitial connective tissue of the lungs (PIE) but also into the pulmonary capillaries in the form of fine bubbles,"' so that cerebral gas embolism results. Blood likewise makes its escape into the alveoli, and bloody frothy sputum may emerge from the mouth and nostrils. If the amount of air escaping into the pulmonary blood vessels is large it gets churned into a frothy mass in the left side of the heart and the victim succumbs rapidly. The important thing for us is that PIE is present, and that it is produced 354 Pulmonic Interstitial Emphysema very rapidly. We must assume alveolar-base rupture, since that is the only way air can possibly leak from the lung air spaces into the connective tissue unless it be through the walls of the bronchi and bronchioles, and there is no evi- dence of that. It should be frankly admitted that we do not know the details of the morphological changes going on in the interior of the lungs in any of these natural overstrain cases; and it may well be that, inider these abnormal conditions, certain areas of the lung are likely to expand unduly and thus to suffer alveolar-base rupture. Accordingly, there may be, even here, local areas of overinflation. If this is so then the mechanics of PIE production would be comparable to those obtaining in the local alveolar-base overstrain cases. The literature on submarine escape and analogous events is growing, and only a little of it has been here referred to. We may confidently look for more enlightenment on obscure points in the near future and, as knowledge un- folds, the role of functional anatomy in facilitating progress in pathological science will be increasingly useful. The important practical point is that air should be diagnosed whenever present and released by aspiration whenever it is causing airblock. Artificial Overstrain Cases.— \x. is always difficult to classify the items in a new subject like the pneumonopathies herein dealt with, and it is quite pos- sible that the PIE resulting from sudden and unorthodox submarine escape should appear under the heading of the artificially induced cases rather than those arising from "natural" living conditions. However that may be, it is certain that the lung-blast cases belong in the artificial group. Lung 5/rt5f.— Although people have been experiencing serious injury and death from explosions for a long time, yet it is only recently that attention has been given to the changes in the lung shown by these catastrophies. There are already many articles on the subject to be found in the literature, but here we shall be able to refer to only a few of them. Although the effect of bomb explosions on the lung (as on the body as a whole) varies from total disin- tegration to relatively slight injury, we are most of all concerned with those cases which survive and in which there is a chance of doing something to aid recovery from PIE and its sequelae. That PIE is caused by lung blast there is no doubt,"* for autopsy findings in the victims show it. There are, too, clinical observational signs in human subjects and X-ray pictures in experimental animals which point to its existence, and syndromic recordings typical of air- block are already to be found in the literature. Lung blast is a special phenomenon composed of two phases, compression and decompression. The time of each phase is very brief, and in this respect the etiology differs from other cases of PIE that we have been reviewing. The severity of the injury varies, of course, with the violence of the explosion and its proximity. For an analysis of its physics, etc., reference must be made to other sources. The effect of the compression wave is like that of a sudden, sharp, but nonpenetrating and diffuse blow struck against the chest wall."" It does not enter via the trachea.™ The ribs are not usually fractured. The intra- Macklin and Macklin 355 alveolar pressure would be raised during this phase. The action is so rapid tliat there woidd not seem to be time for local inetjualities of distention of alveoli to develop, nor could Factor "B" gain head for the same reason. The effect of the decompression wave, which follows immediately after, would tend to convert the hyperatmospheric intrapulmonic pressure to hypoatmospheric, with distention of the alveoli; but it is impossible to visualize the inert lung substance in its reactions to these almost instantaneous happenings. Most of the reports have featured numerous areas of hemorrhage, not only on the sur- face but also in the depths of the lung, as the salient pathological picture. In some ways this suggests a bruise of the lung. We have found no reports on such lungs which have been specially fixed by injection of the preserving fluid into the bronchial tree and which have been studied both grossly and microscopi- cally in an effort to find the exact points where the hemorrhages occurred. It seems clear, however, that the hemorrhagic points were multiple. There has naturally been considerable discussion as to whether it is the compression or decompression phase which is responsible for this damage, with Ztickerman,™ favoring the former. \V^hen reports of lung blast first began to appear in this present war we read the descriptions very carefully to ascertain whether PIE was occurring in the victims who lived for some hours or days after, and our expectations were realized by such reports as those of Dean and his co-workers" and of O'Reilly and Gloyne." We had felt, from our study of the functional anatomy of the lung, that PIE might be predicted in these cases. Although it was not likely that much air would be driven into the interstitial tissues of the lung dining the actual blast waves, yet ruptures would probably then occur through which air would continue to leak into the interstitium and possibly set up an air- block. The accounts which finally confirmed our belief that PIE would occur in some of these lung-blast patients gave the clinical picture as follows, al- though not all of the symptoms might be present in any one case exhibiting any of them: cyanosis of marked degree; extreme dyspnea, the chest being fixed in a position of maximal inspiration; substernal pain; acute abdominal pain for which no explanation could be found at exploratory laparotomy; extreme restlessness requiring opiates; diplopia; and occasional pneumo- thorax. At autopsy, air was found in the mediastinum and pleural cavity, and there was extravasated blood in the carotid sheaths and mediastinum for which no bleeding point could be found, in addition to the numerous areas of hemor- rhage throughout the depths of the lungs. We think it reasonable to postulate that PIE and its sequelae account for these findings. The cyanosis, dyspnea, and chest fixation we have already ex- plained as indicative of airblock. The substernal pain might well be caused by the pressure of trapped air on the mediastinal contents, as in Hamman's*' cases. The abdominal pain might justifiably be attributed to extreme ab- dominal distention with tension on the ureters and pressure on the intestines by retroperitoneal air. Cases have been reported in which acute abdominal 2^6 Pulmonic Interstitial Emphysema symptoms calling for surgical interference have occurred in patients with pneumothorax.^'^ The air might have ruptured into the abdominal cavity in these lung-blast victims, to escape at the time of operation, leaving little or no trace of its former presence. The extreme restlessness and diplopia could be accounted for by cerebral air embolism, which may accompany PIE in severe cases and which originates from air invasion of the pulmonic capil- laries due to augmented intra-alveolar pressure. The blood in the mediastinum may have entered with air from the connective tissue of the lung, and the blood in the carotid sheaths may have similarly invaded them along with air from the mediastinum in an extension of the mediastinal emphysema. It has been found in experimental animals^^ that blood, as well as air, could enter the vascular sheaths of the lung when intra-alveolar pressure was high, and this pointed to an actual fracture of the capillaries of the alveolar bases. Con- tinuing to enter the interstitial tissue with the air, it would soon find its way to the mediastinum, although this last was not actually observed in the ex- perimental animals. From the mediastinum, blood, as well as air, could rup- ture into the pleural cavity. Numerous clinical reports of hemopneumothorax, in which no point of escape for either air or blood could be found, relate to patients in whom there was every possibility of there being a PIE.'*"'^ When pulmonary capillaries rupture, the blood is likely not only to escape into the vascular sheaths and other parts of the interstitium, but also into the alveoli, thus explaining the numerous areas of hemorrhage into the lungs without there being any evidence of external hemorrhage. The presence of blood in the alveoli and in the extrapulmonary vascular sheaths, and of air bubbles within the blood vessels and heart, occurs, ob- viously, only in cases in which the pressure gradient has been high, and in this respect there is an analogy between the victims of lung blast and those of submarine-escape accidents. The decompression phase of lung blast has some similarity in physical conditions with the submarine decompression; but in the latter the time interval is much longer and there is opportunity for Factor "B" to develop. Where pressure variations are less marked, and particularly where they occur less suddenly, although air may escape into the connective tissue, the pneumarrhage is not complicated by hemorrhage. We feel*^ that the lateral roentgenograms of the chest of a cat which had been subjected to lung blast"" give assumptive evidence of air in the anterior mediastinum, as well as of splinting of the lungs, as shown by elevation of the ribs. Others of the roentgenograms and illustrations in this article of Zucker- man's'" support, for us, the idea of PIE being present in the experimental lung-blast victim. The possibilities of airblock and its operative relief in these lung-blast cases has recently been raised,"^ and X-ray examination, particularly with lateral and oblique projections, enjoined. The perhaps more familiar crush injuries of the chest,"" while not so spec- tacular as lung blast, may also show PIE. While these are likely to occur at any Macklin and Macklin 35 y time, they are particularly prevalent in time of total war like the present when people are, for instance, buried beneath the debris of bombed buildings. Even if the air has entered from a wound, rather than from an alveolar-base leakage, it may nevertheless be expedient to remove it; and in any event the physician should be aware of its existence whenever present. We feel that the discussion of this subject of lung blast should not be closed without favorable comment upon the work which anatomical laboratories, like that of the University of Oxford in England, are doing to advance our knowledge. Zuckerman's™ experimental findings made there have been very helpful. As modern war injures humanity, medical science alleviates and heals. We hope to learn much more about lung-blast lesions in the near future. Intratrocheal Anesthesia and Artificial Respiration PIE Cases.— 'Lts^ violent in mode of origin, and more widely distributed, are the PIE cases occurring from artificial alveolar overstrain during the administration of an anesthetic by insufflation, or the induction of artificial respiration, where the pressure within the pulmonic alveoli is too high. This hazard from intratracheal anes- thesia is probably not as great as formerly on account of the safeguards now in use in the improved types of anesthesia machine. The lung must have a con- siderable amount of resistance to alveolar air pressure, otherwise this method of anesthesia could not be used at all. The literature contains accounts of air in the mediastinum or subcutaneous tissues under these circumstances,""** which point indubitably to an initiatory PIE; and of similar meaning is the report of large postrenal blebs of air found in an operation on the kidney." It is not unusual to see descriptions of subcutaneous emphysema following operations in the neck region,*" such as thyroidectomy,^ tracheotomy, or tonsil- lectomy.^** The presence of air is usually ascribed (we think, erroneously, on the whole) to laceration of the airway in the course of the operation, or to nicking the dome of the pleura if the operation was in that region, or to the sucking of air into the wound with inspiration if the operation area involved the mediastinum. It would seem probable that the reason for the frequency of the manifestations of PIE sequelae in operations in the neck region and kidney*" is that insufflation anesthesia is then often employed. Any complicat- ing disease, such as bronchojMieumonia, with atelectasis, ^vould, we think, in- crease the probability of PIE. Operations on the open thorax, when artificial respiration and insufflation anesthesia are used, would seem specially liable to set up an air leak. Marcotte and his co-workers*" have recently pointed out the danger of "positive-pressure" anesthesia in such operations, saying that when the chest has been opened and the pressure within the bronchial tree is raised emphysema occurs more readily; and that when emphysema has once made its appearance the pressure required for its further development is somewhat less than that needed to start it in the first place. They issue the warning that such anesthesia may result in serious complications and even death. The results of experimental overinfla- ti(m of the fresh lungs of calves help to explain iliesc (nidings.''' 3^8 Pulmonic Interstitial Emphysema Closely analogous are the resuscitation cases, when air or oxygen is forced into the lung under pressure. There is always danger of PIE in such induction of artificial respiration. The pulmotor examples will readily come to mind. Cases such as near-drowning, electric shock, gas asphyxia, and asphyxia neona- torum are so often in extremis that desperate measures may be warranted. The airblock syndrome is not manifest, since breathing is almost, or quite, sus- pended. Factor "B" would be a likely contributor to PIE production in these cases. Withdrawal of the air should be done in all of these cases whenever it is deleterious; indeed the idea should be considered by physicians, surgeons, and anesthetists of inserting a hollow needle routinely into the anterior medi- astinum to draw off any air which might happen to have invaded that region during the emergency. In this way we could be fairly sure that the circulatory and respiratory systems would not be incommoded by the treatment. Miscellaneous Cases of PIE There is a group of cases in which air was found in the mediastinum, pleural cavity, subcutaneous tissues, etc., but in which no cause of PIE is obvious. Signs of aberrant air are found, for instance, after the person awakes from sleep, or following a quiet walk, or while sitting in a car, or during shaving, etc. The relatively quiescent nature of the bodily state at the time of the first symptoms has been stressed. Superficially, it would appear that none of the predisposing circumstances were present. When it is recalled, however, that little appreciation exists of the wide variety of the factors which might pre- cipitate PIE, and that considerable time may elapse between the initial causal event and the accumulation of enough air in the mediastinum to be diagnosed, it will be admitted, we think, that in some of these patients an initiatory event might have been found had the physician known what to inquire for and made an honest effort to find it. We have referred to the gravity of atelectasis'" as a predisposing cause in such cases. This is commoner than many physicians think. More frequent and better X-ray examinations in all such cases will, we feel certain, disclose it oftener. Associated infections may well contribute to the likelihood of air rupturing from the compensatorily overinflated regions of lung. An atelectasis, too small to be diagnosed in the usual course, might be large enough to do the damage. We know little or nothing about the relative degree of distention, at any one time, of the various regions of even the normal lung, and it is very diffi- cult to go about finding whether the lung tissue is always uniformly expanded or whether the various parts are always more or less unequally expanded with respect to one another. Many aspects of the related pathology of the lung are no clearer. There is still plenty of opportunity for functional anatomy to aid pathology in the lung. Then, too, the history of these individuals in the period prior to the first appearance of evidence of PIE should be searched by the attending physician. Macklin and Macklin 359 The patient, for instance, might have been lifting a heavy weight, or straining at stool, or violently coughing, even some hours before the fust symptoms ap- peared. It takes time for the air to work its way up the sheaths of the vessels to the hilum of the lung, and more time for it to gather in the mediastinum in sufficient quantity and under sufficient pressure to produce symptoms. Again, prolonged pressure upon one area of the thorax may induce relative collapse of the lung, as in lying on one side too long, or in sitting crouched over the wheel of a car in a long drive, with possible loss of thoracic-muscle tone.*^ Thus in every such case the stage may have been set for local or general alveolar overstrain and a resulting PIE. As physicians become more cognizant of the conditions which usher in PIE they will incpiire more fully into the happenings of the preceding 24 or more hours, and will, we feel confident, find more often a satisfactory explanation for PIE. Things like PIE never "just happen"; there is always a cause. Factor "B" should not be forgotten. There may be doubt that it alone could induce PIE, but in combination with Factor "A" and elevated intra-alveolar pressure it is looked upon as an adju- vant. There is a report of pneumothorax in a patient with a pulmonary em- bolus,-* and it is possible that Factor "B" was here one of the causes of a PIE. Rupture Sites A final word is in order on the sites of air leakage and the nature of the tissue in which the ruptures occur. The lung has been referred to as an air "container" which is efficient under ordinary circumstances. It may be regarded as a special part of the organism built around a little of the gaseous environment, from which it appropriates energizing material, and into which it eliminates certain catabolites. By means of its bellows action the lung is enabled to take in por- tions of this oxygenated gas as needed, and through its specialized structure to subdivide these into minute parcels in order that the surface may be greatly increased. The inner surface of the container, to which this air surface is applied, is similarly widespread, and is essentially a great capillary net or curtain, enormously reduplicated. The ultimate functional anatomical units are the alveoli, which are small contiguous diverticulae from the terminals of the respiratory part of the airway, known as the alveolar sacs. Each alveolus is like a little box with one side left out. Opposite to this opening, or "mouth," is the base. Now there are tioo categories of base. Many bases are partitions between the alveoli of two separate but approximated sacs. Their capillaries function on each of their two sides. Air, in breaking through such a "partitional" base, would merely pass into an adjoining alveolus. An analogous transit of air apparently occurs normally, through the pores. The other type of base does not act as a partition between alveoli; on the contrary, it rests on connective tissue, as already pointed out. Such bases unite to form tubes around blood and air vessels, and layers flanking interlobular septa or underlying the pleura, as already described. Their capillaries, of course, as far as external respiration 360 Pulmonic Interstitial Emphysema is concerned, are functional on only one side. It is these "one-sided," "non- partitional," "sessile," or "marginal" bases that concern us, for air, in break- ing through them, passes, not innocuously into neighboring alveoli, but dangerously into the underlying connective tissue. The nature of the lining of the alveoli has long been a moot problem, and it is important for us here, for it is in this lining that the rupture openings occur. A basic layer of fibers and ground substance is admitted by all. One modern view^ recognizes in the lining a few surviving epithelial cells or "epi- cytes" in the capillary meshes, but in ordinary histological sections no epi- thelium is visible on the "bare areas" which make up most of the total area of the alveolar wall and which overlie the capillaries. Both epicytes and bare areas have been marked out by silver lines, the former more clearly. The final point at issue is the nature of the air-protoplasmic interface in these bare areas. There are some who hold that it is a thin, ordinarily invisible derivative or extension from epithelium; and there are others who think that it is a speciali- zation of mesoderm. It is through this thin film that diffusion of oxygen and carbon dioxide takes place, so that it may be regarded as a vital tissue. It seems to be through this film that minute bubbles of air pass into the capillaries to form emboli when the intra-alveolar pressure is too high. Ordinarily, of course, no leakage of air occurs in it. More work is needed as to the exact situation, in the marginal alveolar bases, of the rupture sites through which minute air freshets pass to cause PIE. It is clear that, if such a base be abnormally spread out, then the fibers of connective tissue in it are separated and the ground substance thinned so that it will be weakened and may become air-pervious. These bases are not built to stand air pressures much above that of the atmos- phere. More experimental work needs to be done on pressures required to rupture these bases in the different laboratory animals. The breaking strain seems to be less in young children than in adult human subjects. In the experi- ments the break-through made itself apparent suddenly, as though many alveo- lar ruptures had occurred simultaneously; we are dealing with Factor "B" here, as well as Factor "A." Once the ruptures are formed the air continues to leak into the connective tissue through them and increase the PIE. When the intra-alveolar pressure is high the capillaries of the base may apparently become torn. Prevention, Diagnosis, and Treatment At first blush it would seem futile to seek to prevent PIE, for it arises so fre- quently under natural circumstances; but a survey of possibilities indicates that something can be done. As with any other ailment, we should try to re- move the causes. The artificially induced overstrain cases would seem pre- ventible, and the avoidance of overly high intra-alveolar pressures, for instance in intratracheal anesthesia and artificial respiration, will suggest itself. The dangers underlying the various predisposing activities, such as heavy lifting, straining at stool, and violent and continuous coughing, should be pointed out to patients by clinicians wherever possible, particularly if there is a history Macklin and Macklin 361 of a previous attack of any pneumonopathy engendered by PIE, such as pneu- mothorax. It is likely, too, that at least some of the pulmonic diseases favoring PIE can be sidestepped, or their dangers lessened, by proper health measures. The prompt removal of foreign bodies and exudate plugs from the bronchial tree, where possible, is particularly enjoined. But, after all this has been said and done, many cases will still occur. What then? It is most important that the physician should become airblock "conscious." Since such signs and symptoms as dyspnea and cyanosis are so frequently as- sociated with primary cardiac incompetence, it is but natural for the physician who finds them present to make his diagnosis "heart disease" of some kind, and let it go at that. When he does this, all opportunity to relieve an existing air- block by depneumatization goes by the board. The physician should always think of the possibility of airblock whenever his patients have such things as dyspnea, cyanosis, inspiratory thoracic fixation, subcutaneous emphysema, tympanites, pneumothorax, anginoid pains, or the peculiar "crunching" sound in the precordium— or, indeed, whenever any predisposing condition is present, particularly atelectasis. An X-ray examination, especially by the lateral pro- jection, should be done as soon as possible to reveal air-pockets,— above all, those of the substernal region. PIE and mediastinal emphysema may be pres- ent, however, when no air has yet reached the anterior mediastinum. Air is most conveniently attacked when it is trapped in the anterior medi- astinum. The technic, by aspiration through a hollow needle, has been made clear by Gumbiner and Cutler,^ and has been particularly successful in the very young infant. For air in the posterior mediastinum the method which Tiegel employed, of suction applied to an incision in the root of the neck, may be used.- Withdrawal of air from this site by way of a catheter insinuated into the mediastinum might also be done when indicated. Summary A knowledge of the functional anatomy of the lung and connected parts, gained by the experimental method, has enabled us to understand many salient features of pulmonic interstitial emphysema and its sequelae occurring in the course of disease and adverse physical conditions. We have found that the anatomical weak spots of the air-container system are the "non-partitional" alveolar bases which lie upon connective tissue. Particularly important as po- tential air-leakage points are those bases which enclose the terminals of the pulmonary arteries and veins. Under adverse conditions these bases are over- strained and are ruptured when the alveolar air pressure rises beyond their powers of resistance. The leak-promoting overstretching of the perivascular bases is seen as an exaggeration of a normal change in them during inspiration, in the course of which the caliber of these vessels is increased to facilitate blood flow through the lungs. We feel that this conception of the usefulness of in- spiratory stroma-pull has not been sufficiently appreciated by physiologists whose attention seems to have been focused on the capillaries rather than the ^62 Pulmonic Interstitial Emphysema arteries and veins. Ruptures may also occur in the paraseptal and subpleural alveolar bases, but have never been found in those around the airway. Various features of these ruptures have been discussed. Atelectasis, by inducing com- pensatory alveolar ectasia, predisposes to air leakage. Interalveolar pores tend to prevent atelectasis by allowing air to diffuse into threatened parts whose bronchus is plugged, and also to minimize the degree of stretching of the af- fected alveolar bases by widening the areas of periatelectatic alveolar ectasia, and so appear to have some influence, though often ineffective, in warding off PIE. Once these multiple minute rupture openings have been made, the likeli- hood is that air will continue to leak into the pulmonic interstitium. The bubbles become larger through coalescence, and most of them move on to the mediastinum, but some remain in various parts of the connective tissue where they interfere with normal lung movements. The conception of the opening and closing action of the extensible pulmonary broncho-vascular "framework" or "skeleton," essential to lung inflation and deflation, enables us to see how, when this action is interdicted by filling of the interstitium with air bubbles, the lung is splinted or locked. Until this happens, the lengthening, and espe- cially the shortening, movements of the broncho-vascular rays may assist in the migration of the bubbles, which are impelled by more recently invading bubbles continuing to be driven through the ruptures by augmented pressure heads in the affected alveoli, as in coughing. The pulmonic connective tissue is seen as a funnel that gathers in air bubbles from the periphery and leads them to the root, from which they debouch into the mediastinum. The his- tological character of the connective tissue explains the preferred courses of the air burrows. A most serious blocking of the pulmonary arteries and veins, by impingement of air bubbles in their sheaths, is indicated in microscopic sections made by approved methods. Pneumomediastinum, or mediastinal emphysema, is a direct consequence of PIE. The biomechanics of pressure elevation beyond that of the atmosphere are considered, and emphasis is laid on the pernicious effect of the air bubbles here on the circulatory mechanism. The syndrome may simulate that of angina pectorit.. The aerogenous escape channels are traced, through which pneumatization of the subcutaneous tissues of the neck, head, upper extremity and chest, the opposite lung, and the pleural cavity is effected. The panemphy- sema field has been surveyed. It is noted that in this air canalization the vascular sheaths are followed by preference not only in the lung, but also in the mediastinum and adnexa, the hiatus aorticus being freely used by air in- vading the abdomen. In all of this, a knowledge of the anatomy of working parts is invaluable. The pool of air in the anterior mediastinum gathers from increments following lines of least resistance from the posterior mediastinum. The selection of these lines, traversing the loose connective tissue between the parietal layers of the pleura and pericardium, is explained by the embryology of the parts. This precordial pool is of particular importance clinically for Macklin and Macklin 363 it may be well demonstrated by lateral X-ray films; the effect of its presence may be listened to with a stethoscope, and it may be operatively removed. Airblock is localized mainly, and frequently altogether, in the mediastinum and limg, and involves the pulmonary circulation primarily. It is the com- bined effect of the interference of aberrant air with the circulation of the blood and the respiratory movements of the lung. It results (a) from impingement of bubbles of air on the pulmonary arteries and veins, and upon the heart and great vessels in the mediastinum, which is particularly serious when the pressure in that space rises above that of the atmosphere; and (b) from im- mobilization of large areas of the lung by air bubbles locked in the connective tissue. It is relieved naturally, to some extent at least, by escape of air from the mediastinum into adjoining regions, including the pleural cavity. When the pressure of the extravasated air in the metamediastinal parts, as the root of the neck, or retroperitoneal region, becomes raised, there is interference with the blood flow in these regions, and the airblock field is thus extended. This field may be, therefore, in its most expansive form, subdivided into three parts: pulmonic, mediastinal, and metamediastinal. Airblock is clinically sensed by the airblock syndrome, which includes dyspnea, cyanosis, and in- spiratory thoracic fixation. It may be confused with heart disease. There may be accompanying signs of spread of air beyond the confines of the medias- tinum. Retroperitoneal air may cause abdominal pain, simulating abdominal disease. Airblock may be overcome by the bodily defense powers, the air being absorbed into the blood stream, but, on the other hand, it may weight the scales in favor of death. A special survey of the literature has revealed a multitude of human cases of different kinds which show evidence of PIE and sequelae, with airblock. All of these are regarded as arising from overstrained nonpartitional alveolar bases which are subjected to unbearable pressure. Alveolar ectasia (usually called alveolar "emphysema") is a very potent exciting cause, since it produces attenuation and weakening of these vulnerable bases. Sudden overstrain is regarded as more important than gradual, as in the latter sort there is time for protective circulatory and other compensatory adjustments to be made. Local alveolar ectasia is regarded as more serious than general, as there is, in the former, more scope for excessive stretching of the nonpartitional bases to take place. Increased intra-alveolar air pressure is also an important exciting cause, and may be set up in a number of ways, all, excepting those of environ- mental induction, coming under the heading of strong expiratory effort, par- ticularly when the glottis is closed. The important predisposing cause is diminution in volume of part of the lung, as the remaining part must then become overexpanded, with overstretching of the nonpartitional bases. Atelec- tasis is the most common volume-reducing condition, and may be single or multiple, large or comparatively small. It may follow occlusion of the bronchi from within or from without. The type due to impaction of foreign bodies is serious because of its sudden production. Atelectasis is frequent in diseases 364 Pulmonic Interstitial Emphysema involving the bronchi, such as bronchopneumonia, where the bronchi are plugged by exudate or by swelling of the walls. Certain allergic conditions be- long here. Spasmodic contraction of the bronchial musculature predisposes. The important thing is that compensatory emphysema is thus set up. In- flammatory changes involving the stretched nonpartitional bases may still further weaken them and make them more easily ruptured; and additional impetus in this direction is provided by acts which raise the intra-alveolar pressure, as coughing. Influenza has been found to be very important in in- ducing PIE. Whooping cotigh is serious. Such cases of PIE are regarded as being due primarily to overstretching of the nonpartitional bases, often having deteriorated protoplasm, combined with intermittent elevation of intra-alveolar pressure. An effect similar to atelectasis is produced by diminution of lung volume from the ablation of a lobe or an entire lung, for the remaining part, withotU thoracoplasty, has to become overexpanded. There is here, however, a com- pensatory feature in the increased blood volume in the surviving parts, which acts thus oppositely to what we have termed "Factor 'B,' " and which would in this way tend to offset the action of Factor "A." Occlusion of the bronchi from without may also induce collapse, but this, being gradual in onset in most cases, is not regarded as so provocative of air leakage from the associated compensatory alveolar ectasia as is the case when the onset is sudden. In this class we have pressure from new growths, particularly primary cancer, or from scars, as in tuberculosis, silicosis, etc. Reduction of volume, too, may occur from continuous presstue on the chest wall, as in lying on the side for a long time, with sudden release, setting up an overstrain in presumably over- distended areas of the lung. There are some cases which support this assump- tion. The noninflation cases occurring soon after birth make a special class, where leakage has taken place on account of the relative overexpansion of aerated parts of the lung. This grotip is notable for the brilliant progress already made in diagnosis and treatment. In some cases, considered tmder the heading "miscellaneous," it is difficult to do more than speculate as to exactly how the PIE is brought about; but here, as in all others, the nonpartitional bases must be involved, and the fundamental features of etiology are doubtless the same as elsewhere. There are other types of case showing PIE and its effects in which the primary factor in bringing about the multiple rupture of the nonpartitional bases seems to be a raised intra-alveolar pressure, though there is nothing to rule oiu coincidental overstretching of these bases, at least in some areas. Prominent in this class, at the moment, is lung blast, during the compression phase; and we have given consideration to this condition and made suggestions for its relief. Here there is sudden violent pressure on the chest. Crush injuries of the chest by impaction of solid masses are somewhat analogous. An interest- ing type of PIE formation occurs in apparently healthy individuals in which the only obvious cause is increased intra-alveolar pressure from violent expira- Macklin and Macklin 365 tory efforts, particularly against obstruction. In the discussion, an effort has been made to visualize the vascular sheaths under the influence of what we have termed "Factors 'A' and 'B.' " Factor "B" is important wherever it exists, and seems to be particularly so in those cases of PIE arising after prolonged muscular effort, as in young men who have just run a "marathon." There is a type of case where the chest is suddenly decompressed, causing air in the alveoli to assume a plus pressure relative to that of the environment. Faulty submarine escape is listed here. The decompression phase of bomb blast may also produce overstrain from this cause, but authorities tend to regard the compression phase as the more important. We have attempted to divide the cases of PIE into two groups depending on whether the nonpartitional bases show local or general overstrain, but it is difficult to draw a sharp line here. The important cases arising from overstrain from too high pressure in intratracheal anesthesia and in arti- ficial respiration may be more general than local in base involvement. Hered- ity seems undoubtedly to predispose to PIE. We have found in the literature so much evidence of unsuspected PIE and its sequelae, with airblock, that we are impelled to believe that the condition is much more prevalent than is realized by the medical profession. We there- fore urge that clinicians be on the lookout for evidences of airblock, that they take all possible means to diagnose it, particularly by X ray, and that they relieve it by surgical intervention whenever it is deleterious to the patient. It seems more than possible, too, that preventive medicine can do something to lower the incidence of PIE. We feel that our predictions of the widespread occurrence of PIE and the types of disease in which it was to be found, made on the basis of our knowledge of the functional anatomy of the parts concerned, have been fulfilled, and that the light shed on the condition has already en- abled clinicians to see their way more clearly and helpfully. Much remains to be done, but we feel that the anatomical approach to such problems will in the future, as in the past, lead to valuable kno^vledge. Final Note Anent our emphasis on atelectasis as a cause of PIE, the paper of Wyatt,^^ just published, is of great interest. He finds, in a series of 135 cases of pneumonia in children, that no less than 26, or 19 per cent, had atelectasis demonstrated roentgenographically. Although he does not mention PIE, nor pneumo- mediastinum, as being present in his atelectasis cases, four of the lateral roentgenograms which he reproduces in his article are, in our opinion, sug- gestive of air in the mediastinum, and, inferentially, of PIE. A letter has also just appeared in the Laiicet by E. Montuschi^ on the sub- ject of airblock in lung blast, in which he refers to the technique of L. Con- dorelli"' for entering the anterior mediastinum to remove air. "The approach," he says, "is from the suprasternal notch, downward, with a curved needle slid- ing behind the posterior aspect of the manubrium sterni." Montuschi states 366 Pulmonic Interstitial Emphysema that he has given a more detailed account of the technique in another place,"^ and further that "This route was devised for the purpose of inducing an arti- ficial anterior pneumomediastinum, which permits in selected cases a more precise diagnosis of conditions arising in this region and of pericardial and pcricardio-sternal adhesions. It is evident that when air under tension is col- lected in the anterior mediastinimi it should be easy to relieve the pressure through this route, as there is no difficulty in penetrating into this region in the normal individual." Much has been written on the entry of air into the mediastinum and pleural cavity from wounds of the neck and chest. One of the most recent of these discussions is that of Neffson"" who analyzed a series of cases of tension pneumo- thorax and mediastinal emphysema occurring after tracheotomy. He admits, however, that the mediastinal emphysema may arise in some of these cases from a PIE originating as described in this paper, from rupture of alveoli into the pulmonary interstitial tissue. REFERENCES 1. Macklin, C. C: in Problems of Ageing, ed. by E. V. Cowchy (2d ed.; Baltimore: 1942), ch. 9. 2. Barrie, H. J.: Lancet 238:996, 1940. 3. Gumbiner, B., and Cutler, M. M.: Amei. Jl. Diseases Childr. 61:650, 1941; also Jl. Amer. Med. Assn. 117:2050, 1941. 4. Hamman, L.: Tins. Assn. Amer. Physns. 52:311, 1937. 5. Hamman, L.: Bull. Johns Hopkins Hosp. 64: 1, 1939. 6. McGuire, J., and Bean, W. B.: Amer. Jl. Med. Scis. 197:502, 1939. 7. Smith, A. B., and Bo\vser, J. F.: Radiology 38:314, 1942. 8. Styron, C. W.: New England Med. Jl. 225:908, 1941. g. Wolff , B. P.: .\nn. Internal Med. 13:1250, 1940. 10. Ehrlich, D. E., and Schomer, A.: Radiology 30:471, 1938. 1 1. Macklin, C. C: Brit. Med. Jl. 11:994, 1937. 12. Berkley, H. K., and Coffen, T. H.: Jl. Amer. Med. -Assn. 72:535, 1919. 13. Lister, W. A.: Lancet 214:1225, 1928. 14. Macklin, C. C: Anat. Recrd. 24:119, 1922. 15. Macklin, C. C: Amer. Jl. Anat. 35:303, 1925. 16. Macklin, C. C: Med. Recrd. 143:89, 1936. 17. Macklin, C. C, and Andrus, P. M.: Anat. Recrd. 45:231, 1930. 18. Macklin, C. C: Physiol. Revs. 9:1, 1929. 19. Macklin, C. C: Archs. Surg. 19:1212, 1929. 20. Macklin, C. C: Amer. Rev. Tuberc. 25:393, 1932. 21. Macklin, C. C: Tubercle 14:16,69, 1932. 22. Macklin, C. C: Archs. Internal Med. 64:913, 1939. 23. Macklin, C. C: Med. Recrd. 150:5, 1939. 24. Macklin, C. C: Jl. Michigan State Med. Soc. 39:756, 1940. 25. Macklin, C. C: Proc. & Trns. Roy. Soc. Canada (s.3:sect.V) 34:69, 1940. 26. Bourne, G.: Brit. Med. Jl. 11:313, 1940. 27. Scott, A. M.: Lancet 232: 1327, 1937. 28. Jessup, P. M.: Archs. Surg. 23:760, 1931. 29. Macklin, C. C: Canad. Med. Assn. Jl. 36:414, 1937. 30. Ballon, H. C, and Francis, B. P.: Archs. Surg. 19:1627, 1929. 31. Fisher, J. H., and Macklin, C. C: Amer. Jl. Diseases Childr. 60: 102, 1940. Macklin and Macklin £^67 32. Fisher, J. H.: Canad. Med. Assn. Jl. 44:27, 1941. 33. Macklin, C. C: Anat. Anzgr. 85(lirgzgshft.):78, 1937. 34. Macklin, C. C: Canad. Med. Assn. Jl. 38:401, 1938. 35. Torrey, R. G., and Grosh, L. C.: Amer. Jl. Med. Scis. 157: 1 70, 1919. 36. Bullowa, J. G. M.: Med. Rccrd. 95:346, 1919. 37. Clark, E., and Synnott, M. J.: Amer. Jl. Med. Scis. 157:219, 1919. 38. Massey, A., and Oldershaw, H. L.: Rrit. Med. Jl. I:6i, 1933. 39. Benjamin, B., and Childs, A. E.: Jl. Pediat. 15:621, 1939. 40. Hurzumache, E.. and Pineles, S.: Amer. Jl. Diseases Childr. 62:661, 1941. 41. Dolgopol, V. B., and Slern, M. E.: Archs. Otolaryngol. 31:140, 1940. 42.Ewart, W., and Roderick, H. B.: Lancet 77:1808, 1899. 43. Clerf, L. H.: Ann. Otol., Rhinol. S: Laryngol. 44:364, 1935. 44. Imperatori, C. J.: Ann. Otol.. Rhinol. & Laryngol. 42:923, 1933. 45. Gumbiner, B.: Personal communication. 46. Moorman, L. J.: Amer. Rev. Tuberc. 42:412, 1940. 47.Hurrell,G.: Brit. Med. Jl. II: 16, 1941. 48. Elkin, W. P.: South. Med. Jl. 34:929, 1941. 49. Macklin, C. C: Proc. & Tins. Roy. Soc. Canada (s.3:sect.V) 28:37, 1934. 50. Macklin, C. C: Jl. Anat. 69:188, 1935. 51. Macklin, C. C: Archs. Pathol. 21:202, 1936. 52. Haues, J. B.: Boston Med. & Surg. Jl. 186:528, 1922. 53. G0tzsche, C: Ugeskr. f. Laeg. 95:765, 1933. 54. Wiilinder, B. E.: Act. tuberc. Scandina\ . 10:66, 1936. 55. Hurxthal, L. M.: New England Jl. Med. 198:687, 1928. 56. Nussbaum, F. H.: Brit. Med. Jl. 11:1169, '937- 57. Phillips, P.: Brit. Med. Jl. 1:54, 1938. 58. Young, R. A.: Brit. Med. Jl. 11:699, 'GSS- 59. Calverley, E. J. G.: Brit. Med. Jl. 11:1899, 1902. 60. Graebner, H.: Archs. Otolaryngol. 29:446, 1939. 61. Blumberg, N., and Latowsky, L. W.: Diseases Chest 6:211, 1940. 62. Caldwell, H. W.: Jl. Amer. Med. Assn. 116:301, 1941. 63. Worden, E. M., and Chaisson, A. F.: Canad. Med. Assn. Jl. 41 :237, 1939. 64. Turnley, W. H.: Jl. Connecticut State Med. Soc. 2:357, 1938. 65. Polak, I. B., and Tibbals, C. L.: U. S. Nav. Med. Bull. 28:862, 1930. 66. Brown, E. W.: U. S. Nav. Med. Bull. 29:366, 1931. 67. Polak, B., and Adams, H.: U. S. Nav. Med. Bull. 30:165, 1932. 68. Macklin, C. C, and Macklin, M. T.: Lancet 242:602, 1942. 69. Cooke, W. E.: Brit. Med. Jl. 1:461, 1936. 70. Zuckerman, S.: Proc. Roy. Soc. Med. 34:171, 1941. 71. Dean, D. M.; Thomas, A. R., and Allison, R. S.: Lancet 239:224, 1940. 72. O'Reilly, J. N., and Gloyne, S. R.: Lancet 241:423, 1941. 73. Steigmann, F., and Singer, H. A.: Amer. Jl. Med. Scis. 192:67, 1936. 74. Hopkins, H. U.: Amer. Jl. Med. Scis. 193:763, 1937. 75. Perry, K. M. A.: Lancet 241:829, 1938. 76. Zuckerman, S.: Lancet 239:219. 1940. 77. Ehrlich, S. D.: Med. Rccrd. 86:926, 1914. 78. Eisen, D.: Radiology 31:623, 1938. 79. Heidrick, A. F.; Adams, W. E., and Livingstone, H. M.: Archs. Surg. 4 i :6i, 1940. So. Lilienthal, H.: TJinrncic Surgery, I (Philadelphia: 1925), 250. 81. Stephens, H. B.: Jl. Tiiorac. Surg. 5:471, 1936. 82. .\ckerman, L. \'., and Brickcr. E. M.: Archs. Surg. 43:445, 1941. 83. del Chicca, S.: Illinois Med. Jl. 73:414, 1938. 84. Keen, J. A.: Jl. Lar\ ngol. .^- Otol. 47: 1 , 1932. 368 Pulmonic Interstitial Emphysema Ss.Math^, C. P., and Faulkner, W. B.: JI. Urol. 46:601, 1941. 86. Marcotte, R. J.; Phillips, F. J.; Adams, W. E., and Livingstone, H.: Jl. Thorac. Surg. 9:346, 1940. 87. Henderson, Y.: Lancet 229:178, 1935. 88. Daniels, L. P.: Nederland. tdschr. v. geneesk. 80:1455, 1936- 89. Wyatt, G. M.: Amer. Jl. Roentgenol. 47:864, 1942. 90. Montuschi, E.: Lancet 242:722, 1942. 91. Condorelli, L.: Minerva med. 27:81, 1936. 92. Montuschi, E.: Post-Grad. Med. Jl. 14:216, 1938. 93. Neffson, A. H.: Archs. Otolaryngol. 37:23, 1943. CHARLES EDWARD BROWN-SEQUARD By RALPH H. MAJOR, M.D. From the DEPARTMENT OF INTERNAL MEDICINE THE UNIVERSITY OF KANSAS SCHOOL OF MEDICINE KANSAS CITY, KANSAS CHARLES EDWARD BROWN-SEQUARD BROWN-SEQUARD WHS oiicc dcscribccl by Harvey Gushing as "the Ponce de Leon of our predecessors," who "thought he had in reality found the foun- tain of perpetual youth." While Ponce de Leon failed in his attempt to dis- cover the fabled fountain, he did succeed in conquering the island of Puerto Rico and was appointed by the King of Spain to the office of governor of "The Island of Florida." Brown-Sequard had his disappointments in endocrinologi- cal discovery, yet lie made many important observations in the virgin field of endocrinology and left an honored record in medical history as a gifted and indefatigable explorer. It would seem fitting in this volinne, which honors one Avho has foinid time from his studies in anatomy and in nutrition to carry on epochal work in endocrinology, to sketch briefly the career of this pioneer of an earlier generation. Brown-Sequard did not have at his disposal the well- equipped laboratories of the twentieth century with their wealth of accimiu- lated chemical and physical knowledge. Much of his work was done in hours snatched from long days of arduous medical practice, frequently interrupted by revolutions, coups d'etat and profound political and social upheavals. Yet, like our colleague and friend, he was gifted with a vivid imagination, an in- terest in many fields of medical research and a most remarkable industry, energy, and perseverance. Charles Edward Brown-Sequard was born in 1817 on the island of Mavuitius in the Indian Ocean, some 550 miles east of Madagascar. This island of Mauritius was discovered by the Portuguese, but settled and then abandoned by the Dutch. It was subsequently under French rule from 1715 until 1810, when it was captured by the British. The British still rule the island, although to this day the majority of the population speak French. According to some authorities, Brown-Sequard's father, Edward Brown, was an Irishman from county Galway, but other biographers state that he was a sea captain from Philadelphia. There is general agreement that his mother. Mile. Sequard, Avas a native of Mauritius and of French extraction. His father, the hardy sea captain, died on a cruise before the birth of his son, and his widow presently brought the child to the United States, where he received his early education. From his earliest childhood, he apparently had not one, but two mother tongues, for he wrote and spoke both English and French with equal facility. While most of his scientific work was done in France and he died as a pro- fessor in the College de France in Paris, according to his obituary in the Lancet, "he never lost an opportunity of expressing his desire that he should be looked upon as a British subject." He was, however, in many respects, a true cosmopolitan and, in the course of his life, practiced medicine in Paris. London, Mauritius, Boston, and New York. Brown-Sequard emigrated to France at the age of twenty-one, and began the study of medicine at Paris. At that time Paris was, unquestionably, the [371] ^'72 Brown-Sequard medical center of the world. The tradition of Bichat, Corvisart, and Laennec was very nuuh alive and molded the thought of both instructors and students. Bouillaud was still active, and Louis, Cruveilheir, Trousseau, Magendie and Orfila were attracting enthusiastic students from all parts of the world. Young Brown-Sequard was thrown into the stimulating atmosphere of a medical center which had only recently produced a revolution in medical thought and was impatient to extend still further the boundaries of human knowledge. In 1846, Brown-Sequard received the degree of doctor of medicine from the Faculty of Medicine of Paris. His doctor's thesis was on the "Vital Properties and Functions of the Spinal Cord," a subject which continued to interest him throughout his medical career. In this thesis, Brown-Sequard reported studies on the motor power of the legs of a frog following section of the spinal cord. He found that immediately after section the motor force was zero, or at most, one-quarter or one-third that before the operation, while one hour later it was greatly increased, often twice the preoperative value. Twenty-four hours later, the motor power showed its maximum increase after which there was usually a gradual decline. He also noted that section of one-half of the cord did not destroy sensation in the areas receiving their nerve supply from the lower portion of the cord. Brown-Sequard continued to study the nervotxs system, carrying out numer- ous experiments and studying patients suffering from nervous diseases. In 1849, three years after his graduation, he published in the Comptes Rendus des SeaJices ef Memoires de la Societe de Biologie an observation that brought him instant recognition and fame— the demonstration of sensory decussation in the spinal cord. This phenomenon, since known as Brown-Sequard's syndrome or paralysis and described in almost every textbook of neurology published since that date, is described by Borland as "paralysis of motion on one side and of sensation on the other side after hemisection of the spinal cord." This subject continued the object of Brown-Sequard's study for years, and fourteen years later he summarized twenty-four cases showing this phenomenon, some the result of wounds, others the result of disease of the spinal cord. Brown-Sequard practised medicine in Paris a few years, devoting his atten- tion mainly to diseases of the nervous system, assembling data on his patients, and carrying out physiological experiments. While he was always primarily interested in the nervous system, his restless spirit refused to be confined in this cadre, so that his interests wandered over the general field of medicine. His restlessness of spirit is also reflected in his peregrinations to various parts of the world. His free expression of radical political views is said to have forced him to flee France in 1852 and to find refuge in New York, where he supported himself by the practice of obstetrics, charging five dollars per case for his services. In 1854 we find him practicing medicine in his native Mauritius, but not for long. The following year he is once more in America, teaching the institutes of medicine in the Virginia Medical College. The next year, how- ever, he was back again in his old haunts in Paris. Ralph H. Major ^73 One of his experiments, reported in the Medical Examiner of Philadelphia, August, 1852, was of especial interest: "My friend, Dr. Claude Bernaid, has recently discovered the curious fact, that after the section of the sympathetic nerve in the neck, the face on the same side and more particularly the ear, becomes warmer and more sensible than the other side. The blood vessels are much enlarged and a great many are visible which were not so before the operation . . . "If galvanism is applied to the superior portion of the sympathetic after it has been cut in the neck, the vessels of the face and of the ear after a certain time, begin to contract; their contraction increases slowly, but at last it is evident that they resume their normal condition, if they are not even smaller. Then the temperature and the sensibility diminish in the face and the ear, and they become in the palsied side the same as in the sound side. "When the galvanic current ceases to act, the vessels begin to dilate again and all the phenomena discovered by Dr. Bernard reappear." This experiment was subsequently carried out and published by Claude Bernard, who was apparently not aware of Brown-Sequard's experiments. Since, however, this observation of Brown-Sequard was a natural corollary of Bernard's work, no controvei sy over priority developed. In 1 856 Brown-Sequard published one of his earliest excursions into the field of endocrinology, an article with the title, "Recherches experimentales sur la physiologie des capsules surrenales." In this paper, he describes removal of the suprarenals in dogs, cats, rabbits, and guinea pigs, and concludes that "death is due to the absence of the suprarenal capsules" and that "the functions of the suprarenal capsules are essential to life." The importance of this work apparently did not impress his colleagues, for, some ten years later, when Claude Bernard read his "Report on the Progi^ess and Achievements of Gen- eral Physiology in France," a report praising the contributions of France to the science of physiology, he made no mention of this work. "It is curious," Olmsted remarks, "that he seems to have missed the significance of Brown- Sequard's work on the adrenal glands in 1856-58, choosing to mention his very much less important work on transfusion." In 1858, Brown-Sequard accepted an invitation to deliver a series of lectures before the Royal College of Surgeons in London. This event was one of the landmarks in his career. He discussed in great detail the anatomy, physiology, and pathology of the nervous system. These lectiaes were published in the Lancet. Their success was phenomenal and established the fame of Brown- Sequard in English-speaking countries. The Lancet of March 12, 1859, notes: "We are gratified to be able to say that our prophecy concerning Dr. Brown- Sequard will be fulfilled. We maintained that his course of lectures 'On the Physiology and Pathology of the Central Nervous System' which lately ap- peared in these columns, was calculated to produce a greater effect in a very large portion of the wide field of medical practice than any course of prelec- tions we had ever before published ... It has happened as wc prophesied, for 2 '74 Brown-Sequard we now hear that, through the exertions of Dr. Hughes Bennett, Dr. Brown- Sequard is to be secured to give a course of lectures at the northern capital, whilst Dr. G. H. B. Macleod has been favourably interceding also for a course at Glasgow. But the infection has likewise spread to Dublin, where the eminent physiologist is afterwards to appear, in consequence of the invitation con- veyed to him through Dr. Robert McDonnell. The College of Surgeons of Lon- don, then, will not be allowed to have had all the harvest to itself, although, to be just, we must give it credit for taking the first steps to gather it in." Brown-Sequard also reaped a certain harvest. Encouraged by the success of his lectures, he migrated to London, and was at once appointed physician to the recently established National Hospital for the Paralysed and Epileptic. Here he availed himself of the splendid opportunity to study various types of nervous disease, but unlike many of his colleagues in positions of similar prominence, he did not garner any golden harvest. This was doubtless due not to lack of opportunity, but to lack of interest. On one occasion he was asked to come from London to Liverpool to see a patient, the fee being 200 pounds. Brown-Sequart replied calmly that he would be in Liverpool in a few days en route for New York, that the patient could see him then, and the fee would be five guineas. On another occasion he was offered by a wealthy American a fee of 10,000 pounds to go to Italy to treat his ailing son. Brown- Sequard refused on the ground that he was not the right person to treat the patient. However little pecuniary affairs interested him, he seems to have been happy with his patients and his studies, and was much pleased when, in 1861, he was invited to deliver the Goulstonian Lectures at the Royal College of Physicians. Meanwhile, in 1858, just subsequent to his first lectures in London, he became editor of the Journal de la Physiologic de I'Homme et des Animaux, and continued to write for this periodical during his sojourn in London. The index of this journal from the year 1856 to 1867 lists fifty articles on physio- logical subjects, the majority treating of the physiology of the nervous system. Brown-Sequard's activity in various political movements has been cited by some biographers as the cause of his frequent departures from France, with his inevitable return. His life in Paris did coincide with several abrupt changes in the government— the collapse of the monarchy of Louis Philippe, the estab- lishment of the Second Republic, and then the foundation of the Second Em- pire of Napoleon IIL Brown-Sequard, however, practiced and worked in Paris on five different occasions and the dates of departure from Paris do not coincide, in the main, with any political upheavals. He seems to have been deeply infected with the wanderlust, possibly an inheritance from his sea- faring father. In 1864, he was appointed Professor of the Physiology and Pathology of the Nervous System at Harvard College, and once more emigrated to the New World. Two years later, to quote Samuel W. Francis in an article in The Medical and Surgical Reporter, August 25, 1866, describing eminent American physi- cians, "Dr. Brown-Sequard, having settled in New York with the specific pur- Ralph H. Major ^jr pose of practising in this metropolis, is not only welcomed by the lovers of science, but may, with the strictest propriety, be included in the present series." Dr. Francis relates that he sought Brown-Sequard's views on smoking to which the doctor replied, "I never smoke, and have seen the most evident proofs of the injurious effects of tobacco on the nervous system." When asked what specialty in medicine interested him most, Brown-Sequard answered, "I am chiefly consulted for nervous affections, both functional and organic, but I am not a specialist; and have studied, and continue to study every branch of medicine." Dr. Francis continues: "Dr. Brown-Sequard's general health has been very good, being exempt from many of the affections that flesh is heir to. But a desire to investigate the contents of his own stomach, under different circum- stances, by means of which he could examine the gastric juice, or partially digested food, has brought on a rare affection, which is sometimes seen in man, namely, a persistent merycism, or rumination, when one is forced to chew a second time what has been swallowed. This has existed since 1844, in conse- quence of his having often performed on himself experiments, consisting in swallowing sponges, to which were attached threads; by drawing upon which the sponges were withdrawn from the stomach, containing gastric juice and liquid or liquified food, which he wished to study." Brown-Sequard, however, did not "settle in New York." In 1867, he re- turned to Paris and the next year became Professor of Experimental Pathology in the School of Medicine. Claude Bernard was one of the first to welcome him and made a point of attending his opening lecture. Brown-Sequard did not remain long in Paris as Professor of Experimental Pathology. In 1873 he was again in New York. Finally, however, in 1878, following the death of Claude Bernard, he was named as his successor. He returned to Paris in 1878 and this time he apparently ceased his wanderings once and for all. He remained in Paris until his death in 1894. These years which Brown-Sequard spent in Paris were marked by his usual experimental activity, as reflected in the numerous articles which continued to flow from his pen. He presently devoted most of his time to the functions of the endocrine glands, a subject which had interested him for many years. As far back as 1869, in a series of lectures of the Faculty of Medicine in Paris, he advanced the idea, to quote his own words, "that the glands have internal secretions and furnish to the blood useful if not essential principles." These investigations of Brown-Sequard presently focused upon him the attention of the medical and lay world, causing him to be an object of veneration to some and an object of derision to others. In an article published in 1891, he empha- sized the proof that had been advanced to support his statement in 1869 that the internal secretions of the various glands played an important role in physi- ology. He reviewed the work which proved that the thyroid, adrenals, and l^ancreas all produced internal secretions necessary for health, pointing out that removal of the adrenals caused death, removal of the thyroid j)roduced ^y6 Brown-Sequard hypothyroidism, and the removal of the pancreas produced diabetes mellitus. About this time Brown-Sequard began his own work on the internal secretions of the testes, work which was to cause such a furor in the scientific world. In one of his early publications he points out that the sexual glands have three important roles: one, that of generation; another, their effect upon the nervous centers "which give to man and to woman their physical, moral and intellectual characteristics"; and thirdly, a special tonic action upon the spinal cord and the brain. He expresses, on numerous occasions, his conviction that the weakness of old men depends, in part, on the diminution of the activity of the sperm glands. Having boldly advanced his theories, Brown-Sequard immediately plunged into the fray to prove the correctness of his conceptions. Some of his case reports are remarkable in the extreme. He frequently saw examples of the striking results obtained from injection of testicular extract or even of semen itself. "Young man, tubercular leprosy. After the first injection he noticed a great increase in strength. He walked many miles on foot, in spite of the heat of summer (in the tropics). He could do his work as a worker because, he declared, of these injections. The hair returned in parts where it had long since fallen." "Young man, age 30, confined to his bed with nodular rheumatism, obliged to rest on his back for several months, ankylosed. After two or three injections he turned himself in his bed and put himself at will on one or the other side. He could bend his right leg up until the middle of the sole of his foot was flat on the bed." "A young doctor in Paris injected, with rapid and complete success, his own sperm under the skin of his wife. She had been in bed in a state of extreme weakness caused by hemorrhage. Her strength returned very rapidly. On four occasions at different times the same weakness, due to the same cause, was re- lieved by dynamogenic action of sperm injected under the skin." A skeptical world refused to be impressed. Dr. Brown-Sequard was deterred, however, neither by criticism, nor abuse, nor by the sarcastic description of the testicular extract, as "Brown-Sequard's Elixir." He continued his work and spent more than 10,000 francs distributing gratis samples of his testicular extract for clinical use. Reviewing this period of Brown-Sequard's activity, we are impressed by his lack of discrimination in assaying the value of the clinical trials. However crude and uncontrolled as much of his work was, we realize that Brown- Sequard's premise that the sexual glands had other functions than those of generation stamps him as possessed of a very keen and observant mind. During this period Brown-Sequard also experimented with extracts of other organs. He employed an extract of the thyroid gland in myxedema with great success, recorded prolongation of life in adrenalectomized animals following injections of a suprarenal extract, and described experiments with aqueous extracts of the pancreas. He also studied the physiological effects of extracts of the thymus, spleen, liver, and kidneys, and advanced the opinion that extracts of the Ralph H. Major ^^y pituitary gland will be of great value in the treatment of hypopituitarism as soon as we can "build up the malady constituted by the special symptoms showing the absence of secretion of the pituitary gland"— a service rendered several years later by Frohlich and others. Since the protocols of these experi- ments contain no records of the blood chemistry or of the blood pressure, it is impossible to tell whether he was working with active extracts. There is cer- tainly no evidence that he had an extract of the pancreas which contained insulin, although, judging from his previous communications, he was ob- viously seeking for some such substance in the pancreas. If Brown-Sequard had had at his disposal the fine procedures of the modern chemist, he would have been able to control his own imaginative thinking and also to avoid the derision of many of his colleagues. Fundamentally, many of his ideas were sound, but he had tackled his work without the tools to fashion it, since these tools were, at that time, undiscovered. Brown-Sequard died in 1894, at the age of seventy-seven, having served medicine faithfully for fifty-six years. His contributions, as we have seen, were numerous and varied. The actual number of his publications exceeds five hun- dred, and, while the majority of them relate to the physiology and pathology of the nervous system and to endocrinology, a review of his bibliogi aphy shows numerous articles also on the subjects of hematology, hypnosis, anesthesia, laryngology, experimental emphysema, vitiated air, and antiseptics. His great- est contributions, however, were his demonstration of the Brown-Sequard syndrome, which was anticipated in his doctor's thesis, and his work on the glands of internal secretion, which, while faulty in certain respects, had the most desirable effect of turning the attention of medicine to this much neg- lected field. We can imagine his great enthusiasm if he were among us today and could view the enormous strides that endocrinology has made since his time. THE UNDISCHARGED OVARIAN FOLLICLE By F. H. A. MARSHALL X From the SCHOOL OF AGRICULTURE, CAMBRIDGE, ENGLAND THE UNDISCHARGED OVARIAN FOLLICLE AS IS well known in the matter of ovulation, there are two main categories i\. of mammal, first those in which the mature ovarian follicles discharge spontaneously, usually at or about the periods of estrus, and secondly those in which the follicles rupture normally only as a result of coition or some other sexual stimulus producing an orgasm. The common domestic animals such as the bitch, the mare, the cow, the ewe, and the sow are examples of the first group; the rabbit, the ferret, and the American ground squirrel are examples of the second group. The cat appears to belong to both groups, individuals ovulating spontaneously or only in response to sexual stimulation in about equal proportions. Among the lower vertebrate classes the same physiologi- cal division occurs, but, whereas among birds ovulation generally appears to depend upon the presence of the male, in reptiles, amphibians, and fishes spontaneous ovulation is the rule, though there are some noteworthy excep- tions (MarshalP). With those species of mammal which do not usually ovulate except as a result of coition there are further marked differences in the be- havior of the nonovulating follicles. In the domestic rabbit two types of follicular atrophy are distinguishable. In the commoner type, which with various minor differences is normal for most animals, the liquor folliculi is absorbed, the epithelium is broken down, the ovum becomes flattened and shriveled, and the cavity eventually becomes filled in by a loose ingrowth of connective tissue from the thecal wall. In the second type, which was described by Heape' and Hammond,' there is a large extravasation into the cavity of blood which clots so as to form a red or black spot; this can be seen on the surface, and in the later stages resembles a blood blister in the skin, whereas the other type is visible only microscopically. A newly formed blood follicle is bright red and becomes dark and eventually black only some time after it has been formed. I am indebted to Dr. John Hammond for allowing me to examine micro- scopically a large series of blood follicles; these show that in the process of formation the loose tissue of the theca interna has broken down in many places, the previously engorged blood vessels having undergone rupture. The rush of blood into the cavity may result in much of the follicular epithelium, together with the ovum being swept aside and disintegrated, the ovum be- coming shrunken. The latter may, however, remain surrounded at a little distance by a reduced number of epithelial cells. The theca interna may be- come completely or almost completely stripped away from the theca externa, or parts of it may remain lined internally by follicular epithelial walls, some of which are hypertrophied as though in an attempt to form luteal cells. The sections show also various transitional stages between mature follicles and dis- integrated atrophic follicles. In the later stages as described by Hammond' the ovum has disappeared as well as the epithelial cells, and the blood follicle is ^82 Undischarged Ovarian Follicle gradually absorbed. Granules of black or brown pigment may be detected in the thecal wall and these arc probably derived from the blood pigment in process of absorption. Blood follicles are pnly very rarely found in other species of mammal and are normal in none of them under any conditions. Blood spots, however, are found in the ovaries of the immature mouse as a result of injecting gonado- trophic hormone (in pregnancy urine); these spots occur in the ovarian stroma and, though some may be large and easily seen, others are very minute (Rob- son*); moreover, it is not always clear that they are associated directly with follicles. It used to be supposed that hemorrhagic follicles in the rabbit represented follicles which failed to undergo ovulation owing to the buck being withheld and that such was probably the normal fate of mature follicles after an estrous period had been going on for a prolonged time. This conclusion, however, is now known to be incorrect, in view particularly of two considerations. First, it is now realized that the formation of blood follicles is the result of a sexual stimulus of the nature of an orgasm and may be produced artificially by the injection of anterior pituitary secretion or pituitarylike substances or by electrical or other stimuli which act on the pituitary through the intermedi- ation of the central nervous system. Secondly, it has been shown through the work of Evans and Swezy° that in the rat, guinea pig, dog, cat, and mouse oogenesis occurs rhythmically throughout the whole period of sexual life and in relation to the ovulation cycle. It had already been ascertained, contrary to what was usually supposed, that in the rabbit (Lane-Claypon") and other animals new ova were produced after sexual maturity, and the earlier litera- ture is reviewed by Evans and Swezy. Moreover, Hill and White, ^ and Smelser, Walton and Whetham® independently discovered that the mature follicle does not persist in the estrous rabbit but is rather short-lived, so that during the prolonged heat periods numbers of follicles appear and disappear, the series overlapping in such a way that at any one time there are approximately the same nimiber of mature follicles on the ovarian surface and estrus and po- tential fertility are continuous. From what has been said it is clear that the formation of hemorrhagic follicles in the rabbit is of the nature of an at- tempted ovulation and probably occurs normally as a result of a sexual stimu- lus of some kind (for instance, when two female rabbits are kept together) which is, however, insufficient to produce ovulation. In other species of mammals which are known to ovulate only after coition, hemorrhagic follicles are not formed, or at any rate have not been described except rarely or as occurring abnormally. In the American ground squirrel Foster^ has given an account of follicular atrophy of the more usual type in which there is no hemorrhage, and such atrophy appears to be the fate of the mature follicles which do not discharge. The ovum, however, is stated to de- generate before the surrounding epithelial cells, whereas in most mammals the reverse is usually the case. F. H. A. Marshall 383 The ferret also differs from most mammals in that it ovulates only after coition (Marshall") and it differs from the rabbit in that no known exceptions to this causal relation have been recorded or observed. The atrophic follicle first figured in 1904 has recently been fully described by Hamilton and Gould." Atresia may set in at all stages of follicular growth and at all times in the cycle. The earliest indications are degeneration of the epithelium and loss of shape. Hamilton and Gould suggest that the interstitial tissue contributes towards the collapse of the follicle by slow cellular proliferation and the pressure which it exerts. The theca becomes fibrotic and constitutes the scars found later in the ovaries. Unlike that of the ground squirrel mentioned above, the zona pellticida of the ovum may persist for a long time along with some leucocytes which remain in the cavity. This description is in general agreement with my own earlier but less detailed observations. Hamilton and Gould state further that the scars of the atrophic follicles of the ferret differ from those of the sow, goat, and mouse in which there is proliferation of the theca that comes to fill the collapsed cavity with fibrous ingrowth, whereas in the ferret there is comparatively little active proliferation of the theca. There is often, however, a loose ingrowth of connective tissue, which may be seen while the ovum is still visible in the cavity (MarshalP"). No hemorrhage from the theca or any part of the wall of the follicle was ever observed, and this may be interpreted as due to the greater density of the tissue as compared with that of the rabbit. Unruptured follicles of a somewhat different type have also been observed in the ferret in some cases where the animals had been on heat for an ex- ceptionally long time. These show a retention with lutealization of some of the epithelial cells not unlike those of the rabbits described above. The ova in various stages of degeneration could also sometimes be seen within such follicles. An attempt was made to discover if there were differences in the histological appearance of the ovary at different times in the prolonged heat periods. For this purpose ferrets were killed after they had been continuously on heat under natural conditions (that is, without artificial stimulation by light or electricity) for five, six, eight, and twelve weeks. It may be stated at once, however, that there was general resemblance in the ovarian histology at all those stages. Each stage showed large follicles, some apparently ready to rupture, but there were no discharged follicles or typical corpora lutea. There were also follicles in process of degeneration, but in the five- and six-week stages these were small ones which apparently had never grown. In the ferret which had been on heat for eight weeks the ovaries contained some medium-sized atrophic follicles with the degenerate ova surrounded by unenlarged or only slightly enlarged epithelial cells. Big degenerate follicles with lutealization of some or most of the epithelial cells as described above were found only in the twelve-week stage. The observations made were not numerous enough to warrant any very definite conclusions, but they suggest that degeneration of large follicles is 384 Undischarged Ovarian Follicle more characteristic of animals which have been on heat for a prolonged period than that it occurs normally in all estrous ferrets. Such a view is con- sistent with what was found in a ferret subjected to ultraviolet light for thirty-eight weeks, but it is to be noted that in this case no degenerate ova were found in the follicles (Marshall'^. These observations suggest that in the ferret, as in the rabbit, ripe follicles develop in relays and then degenerate if ovulation cannot occur, as described for the rabbit by Smelser, Walton and Whetham,^ but that the process does not go on so rapidly in the ferret. Attempts were made to induce ovulation in the ferret by electrical stimula- tion through the head by the method which produced ovvdation or the forma- tion of hemorrhagic follicles in the rabbit (Marshall and Verney'^). With ferrets, however, no reaction on the ovaries was obtained. These negative results may have been due to the stimulation not being applied for a suffi- ciently long period, since ferrets in copulating normally take a very long time over the process, and if the male is withdrawn after a quarter of an hour ovula- tion may not follow. Or they may have been due to the fact that the electrical resistance of the bones of the ferret's skull is greater than that of the rabbit's, and since the electrodes were external to the cranium less current would pass through to the hypothalamus and pituitary (Harris"). With these considera- tions in mind Harris stimulated the hypothalamus directly with an electrode inserted into the posterior region. As a result at post-mortem 102 hours after stimulation three cystic follicles with extensive lutealization were found in the ovaries, but no blood follicles. Summary (1) Among all the mammals in which the reproductive processes have been investigated the rabbit appears to be unique in having ovarian blood follicles as a frequent phenomenon. (2) Ferrets do not ovulate or develop blood follicles (as do rabbits) when experimentally stimulated, possibly on account of the denseness of the tissue surrounding the follicles. (3) Histological study of feirets' ovaries suggests that in this animal, as in the rabbit, ripe follicles develop in relays in the ovaries during estrus, but that in the ferret the process is more marked towards the later stages of the prolonged heat periods. (4) Appearances suggest also that especially in the later stages of estrus the ripe follicles become atrophic, some of the epithelial cells at first becoming hypertrophied while the ovum remains in the cavity of the follicle. REFERENCES 1. Marshall, F. H. A.: Biol. Revs. 17:68, 1942. 2. Heape, W.: Proc. Roy. Soc. London, s,B 76:260, 1905. 3. Hammond, J.: Reproduction in the Rabbit (Edinburgh: 1935). 4. Robson, J. M.: Recent Advances in Sex and Reproductive Physiology (2d ed.; London: 1940). F. H. A. Marshall 38? 5. Evans, H. M., and Swezy, O.: Mcms. Univ. California 9:119, 193 1. 6. Lane-Claypon, J. E.: Proc. Roy. Soc. London, s.B 77:32, 1905. 7. Hill, M., and White, W. E.: Jl. Physiol. 80: 174, 1934. 8. Smelser, G. K.; Walton, A., and Whetham, E. O.: Jl. Exper. Riol. 1 1 1325, 1934. 9. Foster, M. A.: Amer. Jl. Anat. 54:487, 1934. 10. Marshall, F. H. A.: Quart. Jl. Microscop. Sci. 48:323, 1904. 11. Hamilton, W. J., and Gould, J. H.: Trns. Roy. Soc. Edinburgh 60:87, 1940. 12. Marshall, F. H. A.: Jl. Exper. Biol. 17:139, 1940. 13. Marshall, F. H. A., and Verney, E. B.: Jl. Physiol. 86:327, 1936. 14. Harris, G. W.: Proc. Roy. Soc. London, s.B 122:374, 1937. r MECHANISM OF THE DESCENT OF THE TESTICLE UNDER THE ACTION OF SEX HORMONES By THALES MARTINS From the INSTITUTO OSVVALDO CRUZ, RIO DE JANEIRO, BRAZIL MECHANISM OF THE DESCENT OF THE TESTICLE UNDER THE ACTION OF SEX HORMONES THE ELUCIDATION of the mechanism of testicular descent in the postnatal period has lately made real progress. The use of active hormones has demonstrated that the migiation of the organ into its scrotal seat is unques- tionably controlled by endocrine factors. Engle' has shown that after treatment of preadolescent rhesus monkeys with gonadotrophic extracts testicular migration is hastened, a fact that has been corroborated by a number of later papers, which report the use of both hypophysial and pregnancy-urine extracts and extend the application to other zoological groups, man included. Hamilton," using testosterone esters, was able to induce identical effects in rhesus monkeys. Walsh, Cuyler and McCullagh^ and also Nelson and Gallagher* showed that in hypophysectomized rats androgenic substances are able to prevent testicular changes for some time as well as keep the organ in its normal position. In squirrels, androgenic substances may induce spermato- genesis in seasons of sex inactivity (Wells and Moore^). These results do not allow one to exclude a direct action of the hormones upon the testicle. There is left open for discussion the participation of the gonad in the phenomenon, regarding both the modification of its consistency, dimensions, etc., as well as the influence on its own endocrine activity. In order to eliminate these factors, that is, to solve the paradox of the descent of the testicles in a castrated animal, we performed experiments with small paraffin pellets as substitutes for the testicular parenchyma. In such a way a mechanical "testicle" is maintained with exclusion of every endocrine function of the organ. Hamilton,'' in a study directed toward another end, used the technique of placing loose small paraffin spheres in the scrotum in order to discover whether they exert any influence on the development of the cremaster. Our technique and experimental pinposes are, however, quite different from those of this author. In the present paper, we report further experiments on extirpation of the scrotum and the results with regard to the position of the testicle. A preliminary note on the subject has already been published.' Methods Animals: The operation was performed on 17 rats weighing from 180 to 250 gm., and on 4 rhesus monkeys weighing from 3.5 to 9 kg. Operation: In the rats we always used the abdominal method in order to avoid irritation of the scrotal and cremasteric region. The testicles were ex- posed by a small suprapubic incision, from 1.0 to 1.5 cm. in length. Another 1:389] 390 Descent of the Testicle longitudinal incision was made in the tunica albuginea, measuring about y^ of the length of the organ so as to enable the removal of the entire parenchyma, care being taken to clean up the remnants of the testicular pulp which possibly might have remained adherent. A small ovoid paraffin mass of shape and size nearly equal to that of the testicle was then introduced into the cavity of the albuginea; at last the incision was then sutured with extra fine catgut. The operation was always performed on both sides. On the first days after the operation the substituted organ was pushed back and forth towards the ab- domen in order to avoid adhesions which might easily be formed as a result of the irritation caused by such a foreign body and which, in case of fixing the "testicle," would invalidate the experiment. In the rhesus monkey operation by the abdominal route, in order to spare the inguino-scrotal region, is not possible. In one preadolescent animal with testes still in the canal the operation was performed through an incision of the inguinal wall; in the other three it was performed by way of the scrotum. Extirpation of the Scrotal Pouch: In 5 rats weighing from 150 to 200 gm., of course not subjected to the operation with paraffin pellets, the whole skin of the scrotum was extirpated, including the entire region which normally covers the testicles. In 2 of them the cremaster was also extirpated. Treatment: In 4 rats, injection was begun 3 to 1 1 months after the opera- tion; 5 mg. of testosterone prepionate* in 0.5 to 1.0 cc. of sesame oil were given every 3 to 5 days; in rhesus monkeys, starting from 2 to 4 months after the operation, 10 mg. doses were given at the same intervals. Results Of the 17 rats operated and with artificial testes substituted 1 1 were impaired for experimental purposes by local inflammatory reactions, ulcerations, and irremovable adhesions. In the other 6 the result was favorable, the "tes- ticles" being fairly movable, remaining in the scrotal pouch during the first days, and subject to the same mobility as that which is observed in normal animals. For instance, they responded very well to mechanical and thermic stimulants, ascending to the abdomen when the rats were placed at 6° to 10° C, and descending again when the temperature of the environment rose to 25° to30°C. From 2 to 5 weeks later, as the endocrine secretion of the gonad was lack- ing, the false "testes" ascended to the canal and abdomen, remaining there indefinitely in the animals not subjected to hormonal treatment. Four rats, operated from 3 to 1 1 months before, were employed for ex- periments on the action of injected androgenic products. The effects were relatively rapid; even 5 to 10 days after the beginning of the treatment the "testicles" were seen to have moved towards the scrotal pouch being en- gaged in the canal or in the upper third of the scrotum. Within 15 days descent was completely achieved, the "testicles" remaining thereafter consistently in * We are grateful to the Schering Corporation for the donation of the hormonal material. Thales Martins 391 the scrotal pouch (pi. i, figs, i to 3); after cessation of treatment, they returned to their intra-abdominal position. In the rhesus monkeys the lesults were not so readily demonstrable. After the operation the "testicles" ascended to the upper region of the scrotum or into the canal. After 2 to 3 weeks of testosterone treatment, the beginning of descent was detected; the cord became longer, the organ approached the pouch. In one case there was quite a complete descent, the "testicles" occupying the middle region of the scrotum. In the 3 other animals, they descended 2 to 3 cm., lodging in the lower end of the canal or in the upper third of the pouch. A hanging position, like that observed in the rat, was not attained in the rhesus monkey; but the reasons for this may, perhaps, be merely the results of surgical complications. By digital examination we had the impression of the existence of connective bands which prevented complete descent; in one case post-mortem examination showed in fact the presence of adhesions. The differences in the surgical procedures in the two species must also be remem- bered; in the rat it was possible to make a suprapubic approach, all the tracts and adnexa of the testicle remaining untouched; in the rhesus monkey it was necessary to make the incision in the scrotum or the canal which favored the formation of adhesions which were further aggravated by the presence of the foreign body. Nevertheless, we are enabled to conclude that there was a visible effect of the hormonal treatment, although not so positive as in rats. Extirpation of the Scrotum: In normal rats, that is to say in those not sub- jected to the experiment of substituting the testicular parenchyma with paraffin, we observed that the operation of scrotal extirpation was no hin- drance to the maintenance of practically normal position of the gonads, which distended the adjacent nongenital skin (pi. 1, fig. 4). Discussion The results here reported show clearly that in rats, and very probably in rhesus monkeys, testosterone is able to induce the descent of artificial paraffin testicles. The action of the hormone is, therefore, extragonadic, independent of any functional or morphological modification and of the dimensions or consistency of the testicles. What then are the extragonadic conditions that are necessary for the descent of the testicles? It seems that the most important are the following: permeability of the inguinal rings and canal; length and extensibility of the vas deferens, vessels, and smooth fibers of the spermatic cord; development and relaxation of cremaster and dartos muscles; good development of scrotal pouch; sufficient abdominal pressure. When we look for an active motile agent that might cause displacement in the process of descent, we meet with difficulties. There is no muscular organ which has its insertion on the testicle and another fixed extratesticular at- 392 Descent o£ the Testicle tachment which by contraction should induce the descent of the organ. The gubernaculum testis with two moving insertions seems to play no active role in the descent, in the postnatal period at least. Its Latin name means "helm of the testicle," a denomination which we deem very adequate, as we agree with Deming* in his statement that the gubernaculum "is either a guide or rudder, but not a tractor for the testicle." On the other hand, the experiments here recorded on scrotum extirpation show that even in its absence the agents that impel the testicle can force the extragenital skin to distend. Spiral or circular contractions of the musculature of the canal comparable to those which induce the progression of intestinal contents are a conjectural matter, and not yet proved true. It seems that the most important impelling force to be taken into considera- tion, mainly in animals with a lifelong permeable inguinal canal, is the ab- dominal pressure; in certain cases, gravity may also be a facilitating agent. The most relevant role of the testicular hormone would be to annul the resistances and forces that oppose descent; in other words, to develop the vas deferens, spermatic cord, cremaster, and scrotum, while maintaining the genital musculature at a certain degree of relaxation. If the endocrine secretion of the testes is insufficient or absent, the shorten- ing through atrophy associated with greater contractility or tonus of the accessory muscular organs will necessarily lead to ascent of the testicle. Recent work in our laboratory shows that testosterone exerts an inhibitory action on the contractility "in vitro" of the vasa deferentia, epididymides, seminal vesicles, and prostate glands of rats and rhesus monkeys and the vasa deferentia of cats.""^^ The organs of castrated animals or of those treated with estradiol become more responsive to pharmacologic agents and exhibit spontaneous or automatic contractions. In normal or castrated animals treated with testos- terone, there is no automatism and their reactivity to drugs is much less. Experiments "in vivo" corroborate these results (Martins and Valle"'"); more- over, studies of the excitability of the cremaster, as investigated with chronaxy methods, seem to demonstrate that, in rats, the contractility of this muscle is also under the control of testosterone (Martins and Cavalcanti^**). All these results agree well with the conditions necessary for testicular de- scent, namely the checking of muscle actions which would work against the forces that are intended to draw down the testicle. For these reasons, in our preceding work we have pointed out that descent through hormonal treat- ment, either experimental or clinical, should result from effects, not only morphological but also functional. In certain cases, particularly when the effects are very rapid, we can suppose that the dominant action is the func- tional one, namely that which modifies the contractility of the musculature in question. Moreover, in the papers quoted above we were able to verify the fact that the latency period for testosterone treatment to exert influence upon the contractility of the vas deferens and seminal vesicles is relatively short, in rats 42 to 72 hours. Thales Martins 393 In the clinical case published by various auihois, references are made to surprisingly quick results from hormones in the treatment of testicular ectopy; although they rouse a certain skepticism, it is not impossible that the result really derives from the therapy, as the functional action of testosterone can be manifested within a matter of hours, as experimentally demonstrated by us. In cases of pseudocryptorchidism, as Hamilton calls such spastic retentions, the influence of the hormone must be primarily functional; in fact, Hamilton and Hubert" were able to induce in 12 patients a permanent descent of the testicles, using doses of androgens insufficient to provoke erections and scrotal growth. Such clinical results may be regarded as harmonizing with the physi- ologic data discussed in the present paper. Summary Experiments were performed in rats and rhesus monkeys in which the testicu- lar parenchyma was replaced by paraffin pellets inserted into the tunica albuginea cavity. In rats, some days after the operation, the "artificial testicles" ascend towards the inguinal canal or the abdomen, remaining there; but after injections of testosterone propionate they completely descend to the scrotal pouch at the end of 10 to 15 days of treatment. In rhesus monkeys there also was an appreciable but less pronounced descent, due perhaps to the fact that in this species postoperative adhesions are formed with greater frequency. In a group of rats which was not subjected to the experiment with paraffin pellets but in which the scrotal pouch was completely extirpated, we verified the fact that the testes remained in their descended position thanks to the dis- tension of the extragenital skin. Thus it becomes evident that, in rats at least, the scrotum is not an active factor in inducing testicular descent. The results of testosterone propionate in rats with paraffin pellets show that the action of this hormone can appear without any modification of the con- sistency, size, or endocrine activity of the gonads, as the animals, though having a "mechanical testicle" are castrated. The action of the hormone is, conse- quently, extragonadal, controlling the growth and contractility of the testicu- lar adnexa. REFERENCES 1. Engle, E. T.: Endociinology 16:513, 1932. 2. Hamilton, J. B.: Amei. Jl. Physiol. 1 19:325, 1937. 3. Walsh, E. L.; Cuyler. W. K., and McCuIlagh, D. R.: Amer. Jl. Physiol. 107:508, 1934. 4. Nelson, W. O., and Gallagher, T. P.: Science (n.s.) 84:230, 1936. 5. Wells, L.J. , and Moore, C.R.: Anat. Recrd. 66:181, 1936. 6. Hamilton. J. B.: Proc. Soc. Exper. Biol. & Med. 35:386, 1936. 7. Martins. T.: Cpts. rd. Soc. de Biol. 131:299, 1939. 8. Doming, C. L.: Jl. I'rol. 36:274, 1936. 9. Martins, T., and do Valle, J.: Cpts. rd. Soc. de biol. 127:464, 1938. 10. Martins, T., and Valle, J. R.: Cpts. rd. Soc. de biol. 127:1381, 1938. 11. .Martins, I .. and \alle, J. R.: Endocrinology 25:80, 1939. 12. Martins, T.; Valle, J. R., and Porto, A.: Cpts. rd.Soc.de biol. 127:1385, 1938. 13. Martins, T., and Porto, .\.: Cpts. rd. Soc. dc biol. 127: 1389, 1938. 394 Descent o£ the Testicle 14. Martins, T.; Valle, J. R., and Porto, A.: Pfliigers Arch. f. d. gsmt. Physiol. 242:155, 1939. 15. Martins, T., and Valle, J. R.: Cpts. rd. Soc. de biol. 130: 189, 1939. 16. Martins, T., and Valle, J. R.: Pfliigers Arch. f. d. gsmt. Physiol. 243:243, 1940. 17. Martins, T., and Valle, J. P.: Mems. Insto. de Butantan 14: 109, 1940. 18. Martins, T., and Cavalcanti, T.: Unpublished data. 19. Hamilton, J. B., and Hubert, G.: Proc. Soc. Exper. Biol. & Med. 39:4, 1938. EXPLANATION OF PLATE PLATE 1 Figs. 1 and 2. Genital region of rat D5, 11 months after substitution of the testicular parenchyma by paraffin pellets. In 1: the intra-abdominal position of the "testicles" before treatment is marked with ink. In 2: after 15 days' treatment with testosterone propionate; prominence of the descended testicles is seen in the well-developed scrotal pouch. Fig. 3. Reproductive system of rat, 6 months after implantation of paraffin pellets, well tolerated by the albuginea, which appears smooth and vascularized. Fig. 4. Rat 9, 70 days after extirpation of the scrotal pouch. Testicles fairly descended, distending the extragenital skin. [396] A HEMORRHAGIC STATE IN THE VITAMIN E-DEFICIENT FETUS OF THE RAT By KARL E. MASON From the DEPARTMENTS OF ANATOMY, VANDERBILT UNIVERSITY SCHOOL OF MEDICINE, NASHVILLE, TENN., AND UNIVERSITY OF ROCHESTER SCHOOL OF MEDICINE AND DENTISTRY, ROCHESTER, N. Y. A HEMORRHAGIC STATE IN THE VITAMIN E-DEFICIENT FETUS OF THE RAT* DEFICIENCY of vitamin E in female rats is characterized by death and autoly- sis of the implanted embryo during the second week of gestation, usually between the ninth and thirteenth days. The maternal decidua is at first unaffected but some days later shows regressive changes secondary to the loss of fetal elements. It has been generally accepted that the early fetus succumbs to starvation and asphyxiation, which in turn are secondary to retarded development and rarefaction of mesenchymal derivatives, especially those related to the hematopoietic system (Evans and Burr,^ Urner). The resorptive process can be prevented by administration of adequate vitamin E during the first week or ten days of pregnancy. However, when sterile rats are given borderline doses of vitamin E, as in routine bio-assay tests, variable numbers of early resorptions, dead fetuses in difEerent stages of autolysis, and viable fetuses may be observed in uteri ex- amined at laparotomy or at atitopsy during the last third of pregnancy (Mason^*). It appears that fetuses at certain implantation sites escape the early process of resorption but eventually succumb at different periods during the latter part of gestation. Since there exist all gradations between early resorp- tion and viable fetuses, it seemed that an investigation of the nature and causes of fetal death during later stages of pregnancy might afford a better under- standing of the role of vitamin E in fetal development. These studies have revealed the existence of hitherto undescribed abnormalities of the vascular system, associated with late fetal death, the gross characteristics of which are discussed in the present report. Fetal Death at the Sixteenth Day of Pregnancy Although fetal death has been observed at all stages of late pregnancy, an extensive series of studies concerned with the establishment of new criteria of bio-assay response based upon examination of the uterus at the sixteenth day of gestation (Mason'), and the subsecjuent application of this procedure in assay tests (Mason° and unpublished studies) has afforded an opportunity to examine large numbers of dead fetuses-f- and associated living fetuses at * These studies were begun in the Dcpai tintnt of Anatomy of V'andeibilt University School of Medicine, aided by a grant from the Di\ ision of Medical Sciences of the Rockerfeller Foun- dation, and have been continued at the Department of Anatomy, University of Rochester School of Medicine and Dentistry. I wish to acknowledge my indebtedness to Mr. Roger Terry and to Mr. Robert PfafI, Student Fellows in Anatomy, University of Rochester School of Medicine and Dentistry, for their valuable assistance and collaboration in these studies. f Since all transitions exist between typical resorption sites and sites in which death has just occurred, the term "dead fetus" has been arbitrarily applied only to those implantation 402 Fetal Hemorrhage and Vitamin E this particular stage of development. In more than 1600 rats autopsied or laparotomized at the sixteenth day in routine bio-assay tests, approximately 22 per cent have possessed one or more dead fetuses in utero. As many as four or five such fetuses were sometimes encountered in a single uterus. A preliminary study demonstrated that administration of small amounts of vitamin E as late as the twelfth day readily prevents the occurrence of fetal death at the sixteenth day of pregnancy, and also improves the bio-assay response (table 1). Histologically, many viable fetuses removed from uteri containing dead fetuses and resorption sites were normal in every respect. Other live fetuses, TABLE 1 Effect of Additional Vitamin E, Administered on the Tenth and on the Twelfth Days OF Pregnancy, upon the Incidence of Dead Fetuses at the Sixteenth Day of Gestation No. of Vitamin E concentrate fed on Implantation sites at i6th day of pregnancy Positive assay responses, per cent Aver. \vt. of implan- tation sites rats Jth day, mg. loth day, mg. 1 2th day, mg. Live Dead Res. Live, gm. Dead, gm. 12 ID ID 50 50 50 50 38 72 76 15 0 0 49 23 21 58 90 IOC 1.04 I .09 I . 12 0.78 and dead fetuses in which early stages of autolysis were apparent, usually showed a paucity of blood cells in the vessels of the fetus, yolk sac, and fetal placenta, resembling the conditions noted in earlier fetuses by Evans and Burr.^ There were certain exceptions, however, which suggested that fetal death might be attributable to causes other than decreased hematopoiesis. Furthermore, a study of stained smears of the liver and blood of low-E fetuses revealed no qualitative disturbances of hematopoietic activity. Other organs and tissues appeared normal histologically and showed considerable mitotic activity up to the time of death, following which autolysis of the fetal tissues was rapid. On the other hand, certain blood vessels of the fetus were fre- quently distended with blood cells whose staining reactions suggested stasis, while other vessels appeared normal in size but relatively empty. Occasionally there was evidence of hemorrhage into the adjacent tissues, even in embryos which showed no evidence of autolysis. These findings called to mind hemorrhagic areas previously noted in many fetuses removed from uteri when interest was centered upon establishing cri- teria of response to assay dose based upon the weight of the uterine contents at the sixteenth day (Mason'). These areas, attributed at that time to trauma sites in which the total weight of the placenta, fetus, and intact fetal membranes exceeds 0.4 gm. The usual weight of dead fetuses and their adnexa varied from 0.70 to 1.2 gm., that of associated living fetuses from 0.92 to 1.34 gin. Karl E. Mason 403 induced during the removal and rupture of implantation sites, on closer ex- amination proved to be associated invariably with fetuses which had just died or with those whose continued survival could be considered questionable. In all subsequent bio-assay tests for vitamin E (in which positive responses are based upon the presence of two or more viable fetuses at the sixteenth day of pregnancy) implantation sites containing dead or living fetuses have been routinely examined with the aid of a dissecting microscope. The intact sites were placed in warm physiological saline, the state of circulation in the yolk sac noted, the fetal membranes carefully removed, and the fetus with its at- tached umbilical vessels and placenta examined grossly. The location and extent of vascular injury in each fetus, and the relative position of the latter TABLE 2 The Frequency of Vascular Changes in Dead and Live Fetuses at the Sixteenth Day No. of liv; fetuses No. of dead fetuses No. of assay rats Total examined With vascular injury Per cent Total e.xamined With vascular injury Per cent 210* . 1228 4,^^ Is 186 95 22t * This number includes only rats in which one or more of the living fetuses showed vascular changes. t Autolysis was so far advanced in many of the dead fetuses that no evidence of the vascular damage remained. in the uterine horns, were recorded. The data of table 2 present the incidence of vascular changes observed in 210 rats possessing one or more viable and affected fetuses at autopsy on the sixteenth day. Vascular Changes in Fetuses at the Sixteenth Day The earliest changes observed consisted of dilation and congestion in super- ficial blood vessels. Vascular plexuses in the region of the cranial vault, external ear, shoulder, and dorsolateral portions of the trunk were most fre- quently involved. In the capillary bed related to these plexuses there often were petechia-like areas of variable size and shape which, under high magni- fication, appeared to represent localized areas of extensive capillary dilation and stasis. The gioss appearance of normal fetuses is shown in plate 1, figure 1 . The injected specimen shown in plate 1, figure 2, illustrates the character of this vascular congestion in superficial plexuses of the supraorbital, posterior au- ricular, and shoulder regions, and in more localized areas on the limbs and lateral body wall. The dark color of the latter is due to retained blood, for the ink suspension usually failed to penetrate these larger areas of vascular stasis. The plexiform dilations and petechiae usually appeared in the venous chan- nels, whereas the more extensive areas of hemorrhage to be described later, involved also the arterial trunks and their capillary bed. However, it was often impossible to determine which division of the vascular system was con- cerned. Although normal circulation could be observed in vessels immediately adjacent to the affected areas, as well as elsewhere, the volume of blood was 404 Fetal Hemorrhage and Vitamin E often reduced to such an extent that the fetus appeared distinctly pale. This pallor and the areas of vascular injury could be noted before removal of the fetal membranes. Fetuses in which death was imminent, or had recently occurred, showed more striking and irregular distention of superficial vessels, associated with marked dilation of deeper vessels in the immediately vicinity. Most frequently involved were the large vascular trunks at the root of the neck and their more superficial branches in the lateral thoracic wall near the base of the forelimb (pi. 1, fig. 3). These latter changes were almost universally accompanied by prominent areas of hemorrhage in the lateral wall and floor of the cerebral vesicles. Histological examination revealed extensive extravasation of blood into the substance of the corpus striatum and adjacent regions of the cerebral hemispheres. Similar but less extensive areas of hemorrhage were sometimes encountered in other regions of the brain and in the spinal cord, tail, extremi- ties, and liver. Fetuses such as depicted in plate 1, figure 3, if viable, showed an almost imperceptible circulation in the vessels of the yolk sac and in the super- ficial capillaries of the fetus proper, while that of the umbilical vessels was diminished to a variable degree. Yet, on histological examination, the fetal tissues involved by vascular congestion or hemorrhage as well as those not so affected appeared normal and sometimes showed considerable mitotic activity. Occasionally the terminal stages of fetal life were characterized by more diffuse areas of hemorrhage in the superficial tissues of the head and body wall (pi. 1, fig. 4), in addition to certain of the vascular changes described above. Sometimes large areas of superficial edema in the body wall and a bloody tinge to the amniotic fluid were noted. It seemed incredible that fetuses with such extensive vascular injury and edema could still be viable. The striking pallor and vascular changes of fetuses such as pictured in plate 1 , figures 3 and 4, could be noted before the uterine wall was opened. Although similar changes were found in fetuses which proved to be dead on examination, they were soon obliterated by autolytic changes in the fetus. A single protocol will serve to illustrate the diverse conditions observed in fetuses at autopsy on the sixteenth day of pregnancy. Protocol Rat 3697 received 1.75 mg. of synthetic beta-tocopherol, distributed over the first ten days of pregnancy; a dosage level Avhich gave a positive response in six of eleven rats tested. At autopsy there were eleven viable fetuses, six in the right horn and five in the left horn of the uterus. The finding in individual fetuses, beginning at the cephalic end of the uterine horns, was as follows: Ri — Normal R2 — Congestion in superficial vessels of shoulder region R3 — Superficial congestion near ear, shoulder, and lateral body wall R4 — Superficial congestion in extensive plexus lateral to lumbar and sacral vertebrae R5 — Moderate congestion over shoulder and hip region; deep hemorrhage in right cerebrum R6 — Congestion in deep vessels overlying spinal cord in upper lumbar and thoracic region, and in superficial vessels around eye Karl E. Mason aq^ Li — .Superficial congestion aioinul right shoulder and over both parietal regions L2 — Normal L3 — Superficial congestion in posterior auricular and left parietal regions L4 — Superficial congestion in right shoulder and o\er lumbar vertebrae Lr, — Extensive congested plexuses over dorsum of trunk and head. Multi]jlc petechiae overlying cervical and thoracic spinal cord, with distention of vertebral vessels on each side; other jjetechial spots over cerebrum and cerebellum The vascular changes described are unquestionably pathognomonic for vitamin E deficiency. Their incidence, and that of late fetal death, have con- sistently proved to be inversely proportional to the magnitude of borderline levels of giaded dosage in bio-assays of the various natural and synthetic forms of vitamin E. It has been conclusively demonstrated that the daily administra- tion of relatively large amounts of ascorbic acid or vitamin K (2-methylnaptho- quinone) during the first sixteen days of pregnancy does not influence eithci the occurrence or the character of the vascular injury. Furthermore, vitamin E-deficient fetuses have been found to contain a normal content of vitamin C as measured by the silver nitrate technique, and rats suffering from pro- longed deficiency of vitamin E have shown no alteration in blood-clotting time.* Discussion The ultimate cause of fetal death in the vitamin E-deficient fetus at the six- teenth day of gestation is undoubtedly the same as that operative at earlier stages of development, namely, diminiuion of the circulating blood to a level inadequate for continued survival of fetal tissues. The question of paramount importance is whether this ischemic state is brought about by a suppression of hematopoiesis, as suggested by Evans and Burr^ in their study of younger fetuses, or by a defect in the vascular wall leading to congestion, stasis, and extravasation of blood into the tissues. Acceptance of the first explanation implies that the vascular injury is the natural result of decreased blood vol- ume. Acceptance of the second alternative implies that the evidence sugges- tive of impaired hematopoiesis is really an artifact caused by vascular dilation and congestion associated with hemorrhage into the soft tissues. It cannot be denied, however, that both processes might be operative at diff^erent periods of the development or even simultaneously. The observations presented here afford some evidence that the hemato- poietic activity of the liver at the sixteenth day of development may not be appreciably depressed in fetuses exhibiting early vascular lesions. However, the variable picture encountered in different fetuses and the rapid telescop- ing of the terminal events have so far made it impossible to state with cer- tainty whether phenomena suggestive of hematopoietic dysfunction are the cause of, or the result of, the vascular lesions. Little can be said at present concerning the exact cause or nature of the * I am indebted to Dr. C. P. Leblond for analysis of the fetal tissues by silver nitrate precipi- tation, to Dr. Sarah Hooker and Mr. R. \. Bruce for determinations on the blood-clotting time. 4o6 Fetal Hemorrhage and Vitamin E vascular changes. Since frank hemorrhage occvxrs most frequently in areas such as the cerebral wall, where the tissues are relatively more compact than those in which the vascular channels arc usually dilated and thrombosed, weakness of the extravascular supporting tissues does not appear to be a factor. The occurrence of demonstrable edema only in advanced stages of the disorder sug- gests that altered permeability of the vascular wall is a secondary phenomenon. Undoubtedly, the varying incidence and gross characteristics of the vascular changes in different regions of the fetus constitute factors whose significance cannot be fully appreciated at the present stage of these investigations. There is need for a reinvestigation of the early resorptive process in order to determine whether there exist abnormalities of the vascular system com- parable to those described above. Should such prove to be the case, the histo- pathology of vitamin E deficiency in the female rat would have much in common with spontaneous hemorrhage observed in vitamin E-deficient chick embryos (Adamstone'^' '), together with the capillary thrombi in cerebellar and cerebral vessels (Pappenheimer and Goettsch^) and the exudative diathesis (Dam and Glavind") observed in young and adult chicks deprived of vitamin E. It should be mentioned, however, that the present state of our knowledge does not permit an explanation of the various histopathological manifestations of vitamin E deficiency on the basis of either a structural or physiological de- fect of the vascular wall or a suppression of hematopoiesis. These can be re- garded only as a manifestation of underlying metabolic disturbances which, when finally understood, will afford the real answer concerning the physiologi- cal role and function of vitamin E. A more extensive discussion of these vascular phenomena and their correlation with other lesions of vitamin E deficiency in laboratory animals has been presented elsewhere.'" Summary and Conclusions Fetal death is of frequent occurrence in sterile vitamin E-deficient rats given borderline doses of vitamin E and autopsied at the sixteenth day of pregnancy. Prior to exitus there appear abnormalities in the vascular system character- ized by variable degrees of dilation and thrombosis of peripheral and deeper vascular channels and by either localized or diffuse areas of hemorrhage which are sometimes associated with edema, together with a paucity of blood in other vessels of the fetus and in those of the fetal membranes. The fetus suc- cumbs to generalized ischemia. Although suppressed hematopoiesis in the liver and yolk sac, characteristic of earlier fetal death and resorption (Evans and Burr^), cannot be excluded as a contributory factor, vascular injury is regarded as the primary cause of late fetal death. These vascular changes, which can be directly correlated with the level of vitamin E dosage and are not influenced by separate administration of vitamin C or vitamin K, are undoubtedly pathognomonic for vitamin E deficiency and resemble in many respects the capillary thrombi and exudative diathesis ob- served in the vitamin E-deficient chick by other investigators. Karl E. Mason ^07 REFERENCES 1. Evans, H. M., and Burr, G. O.: Mcms. Univ. California v.8, 1927. 2. Urner, J. A.: Anat. Read. 50:175, 1931. 3. Mason, K. E.: in Vitamin E: a symposium ([Cambridge (Engl.): 1939]), 31. 4. Mason, K. E.: Jl. Nutrit. 23:59, 1942. 5. Mason, K. E.: Jl. Nutrit. 23:71, 1942. 6. Adamstone, F. B.: Jl. Morphol. & Physiol. 52:47, 1931. 7. Adamstone, F. B.: Archs. Pathol. 31:622, 1941. 8. Pappenheimer, A. M., and Goettsch, M.: Jl. Exper. Med. 53: 11, 1931. 9. Dam, H., and Glavind, J.: Skandinav. Arch. f. Physiol. 82:299, 1939. 10. Mason, K. E.: Yale Jl. Biol. & Med. 14:605, 1942. EXPLANATION OF PLATE PLATE 1 Photographs of rat fetuses at the sixteenth day of pregnancy, with fetal mem- branes removed but umbilical circulation intact, x 5. Fig. 1. Normal rat fetus showing appearance of vascular system. Fig. 2. Vitamin E-deficient fetus (fixed in formalin) injected with a suspension of India ink before death, showing dilated vascular channels in supraorbital, posterior auricular, and lateral trunk regions. The ink suspension penetrated the small petechial spots on the extremities and dorsal midline, but failed to enter the larger areas along the lateral body wall which are still filled with blood. The fetal heart continued to pulsate after injection was discontinued. Fig. 3. Vitamin E-deficient fetus showing marked dilation and congestion of vascular plexuses just caudad to forelimb, marked distention of deeper vessels just cephalad to forelimb, and extensive bilateral hemorrhage in the floor and lateral wall of the cerebral vesicles. This fetus was living but showed meager circulation in peripheral capillaries and a marked reduction in blood flow through the umbilical vessels. Fig. 4. Vitamin E-deficient fetus in which faint circulation was demonstrable in vitelline and umbilical vessels despite the extensive localized congestion in certain vessels and the diffuse areas of hemorrhage in other regions. [408] Plate 1 [409] RELATIONSHIPS OF SODIUM AND POTASSIUM TO CARBOHYDRATE METABOLISM By IRVINE McQUARRIE, Ph.D., M.D. From the DEPARTMENT OF PEDIATRICS, UNIVERSITY OF MINNESOTA MINNEAPOLIS, MINNESOTA RELATIONSHIPS OF SODIUM AND POTASSIUM TO CARBOHYDRATE METABOLISM DURING the past few years an increasing number of investigators have called attention to the possibility of interrelationships between the monovalent elements, sodium and potassium, and carbohydrate metabolism. However, no effort has as yet been made to correlate the results of experiments carried out from many different angles and no place has been given to the subject even in the latest monographs or textbooks on physiology. The present review represents an attempt to summarize and evaluate pertinent data from widely scattered sources, special reference being made to reports concerning the an- tagonistic effects of sodium and potassium on carbohydrate metabolism in diabetes mellitus. Relationship of Sodium to Carbohydrate Metabolism The first suggestion that mammalian carbohydrate metabolism might be influenced by the administration of sodium salts was the observation by Murlin and Kramer^ that sodium hydroxide and sodium carbonate increased the utilization of glucose by depancreatized dogs. Murlin and Graver" and UnderhilP subsequently reported marked decreases in glycosuria in diabetic patients following administration of these alkaline salts in liberal amounts. The effects observed were attributed to induced changes in the acid-base equilibrium of the body toward the alkaline side. Subsequent studies on the effects of changes in the acid-base balance on carbohydrate metabolism, however, have been somewhat contradictory. For instance. Beard* reported finding little or no effect from giving moderate amounts of sodium bicarbonate to diabetics but incidentally recorded the fact that two of his patients, who voluntarily took increased amounts of sodium chloride and water, showed increases in their carbohydrate tolerance. He drew no conclusions from these observations. Thompson, Mitchell and Kolb" found in normal subjects that acid administration was followed by diminution of carbohydrate tolerance as indicated by blood-sugar curves, while alkalosis had little effect. In a series of acute experiments on normal children, Johnston and Maroney" found the oxidation of dextrose to be ac- celerated by a slight shift toward the acid side of neutrality but depressed by marked acidosis. It was inhibited on the alkaline side. They also found that the oxidation of dextrose was at first depressed but later accelerated by in- gestion of NaCl and water. It would appear from these conflicting residts that acid-base balance per se plays little or no direct role in carbohydrate metabo- lism. Schenk,^ who investigated the effects of intravenous administration of vari- ous electrolytes on blood sugar, found that the latter was decreased signifi- [413] 4 1 4 Relation of Salts to Carbohydrate Metabolism cantly by injection of a lo per cent solution of NaCl. This change was attributed entirely to blood dilution, resulting from the hypertonic infusion. In an attempt to determine the influence of various electrolytes on the action of insulin, Abelin and Goldener* at about the same time found that 2 gm. of NaH^PO^ or 2 gm. of a mixture of NaCl, NaHCOg and Na^SO^ given by mouth to a rabbit produced marked augmentation of the hypoglycemic effect of 4 to 6 units of insulin injected simultaneously. Semler" reported finding no change in the blood sugar of two diabetic men following the injection of 5 gm. of NaCl without insulin. Because of the finding of low blood chloride in diabetics and high blood sugar in various conditions accompanied by hypochloremia, such as intestinal obstruction or histamine poisoning. Glass and Beiless"" reinvestigated the effects of injecting hypertonic NaCl solutions in diabetic subjects. In ten such patients they found that a single injection of 3 or 4 gm. of chemically pure NaCl in a 15 to 20 per cent solution caused a decrease in blood sugar of from 12 to 43 per cent with great regularity. A 12- gram dose of NaCl given by mouth had a similar effect. In explanation of their results, these authors claimed to have excluded the possibility of blood dilution or of spontaneous fluctuation in blood-sugar concentration. They attributed the blood-sugar-lowering effect to the CI ion rather than to the Na ion and suggested that the salt exerted a vagotonic action. Relationship of Potassium to Carbohydrate Metabolism No reference is made to a possible relationship between K and carbohydrate metabolism in any of the foregoing papers, in spite of the fact that the possi- bility of such a relationship was first suggested shortly after insulin was dis- covered. Briggs and co-workers" found an average decrease of 24 per cent in the serum K of fasting dogs following the injection of insulin in large doses. Shortly thereafter Harrop and Benedict" independently observed in both normal and diabetic subjects that a fall in K accompanied the well-known decreases in blood glucose and inorganic phosphorus, which result from administration of insulin. They suggested that K may enter into the inter- mediary metabolism of carbohydrate in association with the phosphorus- carbohydrate combination. Many other investigators (Wigglesworth and co-workers;'^ Hausler and Heesch;" Kerr;" Cori and Cori;'* Ellsworth and Weinstein;" Keys;'* McQuarrie and co-workers'^) have likewise found insulin hypoglycemia to be accompanied by a simultaneous fall of plasma K and PO^. Data obtained subsequently from many other sources have strengthened the hypothesis that K participates in some essential manner in the synthesis and breakdown of complex polysaccharides both in plant life and in the animal body. Plant cells behave in a general way like the liver cells in animals as regards the movements of K and sugar. Pulver and Verzar^ showed that K and sugar absorption and output are paralleled in yeast metabolism. Day and Comboni^' studied the role of K in starch synthesis by plants (Pisum sativum) giown from seeds in sand cultures with complete nutrient solution Irvine McQuarrie 41^ containing potassium phosphate. When the starch content of plants grown in this medium was compared with that of similar plants grown under identi- cal conditions, except for substitution of the NH^ ion for the K ion, it was found that the plants raised in the K-deficient solution produced but half as much starch. Lasnitski and Szorenyi" reported that the anaerobic fermenta- tion of glucose by baker's yeast increased on the average by about 150 per cent if K (0.01 M KCl) was added to the medium ^vhich contained as other cations NH, and Mg. Lasnitski'^"^ earlier found that anaerobic fermentation of glucose by tumor tissue or normal tissue, especially brain cortex, was markedly gieater in a medium containing both K and Na in physiological concentrations than in a medium containing Na alone. Ashford and Dixon"' and Dickens and Greville™ found that the aerobic fermentation and oxidation of glucose by brain cortex could be increased considerably by the addition of a surplus of K to the medium. That is, K appeared to activate the enzymatic breakdown of glucose. Excepting in conditions like adrenal insufficiency (serum K elevated, sugar low) and hypercorticoadrenalism (serum K at times decreased, sugar in- creased), it is true in general that glucose, inorganic PO^, and K appear to move out of or into the blood together. Fenn"' showed that deposition of glycogen in the liver was necessarily accompanied by the deposition of K and water. The quantity of K so deposited, however, was far less than the quantity of glycogen (measured in equivalents), whereas the blood changes would sug- gest equivalent amounts. Flock and co-workers"^ found that intravenous ad- ministration of glucose or levulose at a constant rate produced a decrease of serum K in normal dogs. The decrease in K, expressed in percentage of original concentration in the serum was not so large as that simultaneously produced in the concentration of inorganic P. Sodium lactate and sodium pyruvate produced similar results, but their effect was attributed to the Na ion because NaHCOg and NaCl were found to be equally effective. In depancreatized dogs, not receiving insulin, injection of glucose had variable effects on the level of serum K. In a later investigation"^ the same authors found that injection of 10 per cent glucose solution, with or without cortin, or 0.9 per cent NaCl solution, caused a decrease in serum K in adrenalectomized dogs. They ob- served that NaCl, like cortin, when given with glucose solution to adrenalec- tomized dogs, prevented the rise in K which occurred if glucose solution alone was administered. NaCl appeared to be the full equivalent of cortin in respect to the excretion of K and the response of the adrenalectomized dog to an injection of glucose. The disposal of injected glucose occurred to the same extent, and the same changes in K and inorganic P of the serum were seen in the adrenalectomized dog as in normal dogs. They concluded that "the prob- able influence of inorganic ions, such as Na and CI, is through osmosis, ion antagonism and changes in the permeability of membranes." Long and co- workers^" likewise reported that the essential chemical changes associated with carbohydrate metabolism could proceed in adrenalectomized dogs without 41 6 Relation of Salts to Carbohydrate Metabolism the "salt and water" hormone of the adrenal cortex, provided an adequate supply of Na and CI ions were available. In partially depancreatized rats NaCl did not produce glycosuria as the hormone did. Pulver and Verzar" offer in support of the theory— that K is involved in the phosphorylation process leading to the formation of a polysaccharide— the observation that substances which retard phosphorylation inhibit the uptake of both glucose and K. "From this point of view," says Lasnitzki,*' "there appear to be at least two main possibilities regarding the explanation of the effect of K on fermentation and oxidation of glucose. Either K activates phos- phorylation and thereby the subsequent formation of the supposed poly- saccharide, or it activates the breakdown of the latter. In the former case, K might be bound chemically, while in the latter case it might be adsorbed by the polysaccharide (probably having colloidal properties). In either case K would be liberated with breakdown of the polysaccharide." Verzar and Somogyi^"'^^'^* suggested that K is organically bound in phosphorylation of sugar. The latter process they regard as being under the control of the adrenals in mammals. The disturbance in glycogenesis resulting from adrenalectomy accounts for there being no increase in output of K to the blood from muscle stimulation in the adrenalectomized animal, comparable to that shown by Fenn^^ to occur in the normal animal. As first shown by D'Silva,^"'^ the concentration of K in the serum of animals is regularly increased for a few minutes after the intravenous injection of adrenalin. Posterior pituitary extract (pitressin, P.D.) had a similar effect, but the rise was more delayed. Both agents are known to increase blood sugar, also. The rise in serum K after adrenalin was shown to be followed by a more prolonged fall, which was attributed to stimulation of insulin production by the hyperglycemia. This "after fall" of K was practically absent in depancre- atized animals. Paralysis of the parasympathetic system with atropine likewise prevented the fall. The temporary rise in serum K after adrenalin injec- tions has been confirmed by subsequent work (Marenzi and Gerschman;''*-^ Schwartz;*" Houssay and co-workers"). In the human subject, Keys'^ found a more transient increase in serum K than reported for lower animals. Houssay, Marenzi and Gerschman" showed that the K entering the blood, following injection of adrenaline or direct stimulation of the sympathetic nerves to the liver, comes from the liver. Snyder and Johnson*^ demonstrated that the output of both K and glucose from the perfused liver of the turtle was decreased by vagal stimulation. Fenn" points out that K and lactic acid appear to move together in re- lationship to muscular activity just as K and glucose rise and fall together under certain conditions apparently concerned with the movement of glucose into or out of the liver. Lactic acid and K increase together in the venous blood plasma when muscles contract as they do in asphyxia, after hemorrhage and in histamine shock. In the light of these observations, Fenn states: "It is tempt- ing to suggest that K leaves the liver or enters muscle in company with glucose Irvine McQuarrie ai^j while it leaves the muscle or enters the liver along with lactic acid. Some sup- port for such a tentative hypothesis is found in the observation that after muscular activity there is a consistent increase in the K content of the liver." The existence of a K cycle from muscle to liver and other viscera, and back again, comparable to the lactic acid cycle of Soskin" and Cori'' is suggested. In this connection it is interesting to refer briefly to studies on the K and carbohydrate metabolism in the clinical condition known as familial periodic paralysis. In this disorder serum K is found to be very low during periodic attacks of muscular paralysis (Walker'"). Attacks are brought on by excessive carbohydrate intake. Administration of K salts raises the serum K and causes prompt relief of the paralysis (Aitken and co-workers"). Aitken and co-workers" showed that paralytic attacks could be induced at will in such patients by any procedure that lowers serum K. The K-lowering effect of glucose or levulose by mouth or by vein was found to be more marked than in the normal subject. Allott and McArdle'^ found by balance studies that K is retained by the body during an attack. Phosphate was the only other inorganic substance found to be altered in attacks. It tended to parallel changes in K. The disease, there- fore, appears to involve the hexosephosphate mechanism in some manner. Antagonistic Effects of Na and K on Carbohydrate Metabolism The attention of the author and his associates (McQuarrie and co- workers*"' ^°) was first directed to the possibility of an interrelationship between the Na and K economy and carbohydrate metabolism in 1930, when they observed freakish variations in the degiee of glycosuria shown by an 8-year-old diabetic girl, who manifested an abnormal craving for salt. Although the patient was main- tained on a uniform diet, it was found that the total amoiuit of sugar in the urine for a given period tended to \ ary inversely with the quantity of NaCl which she was allowed to consume. Unfortunately, an attempt to carry out a more complete investigation of the observed phenomenon in that patient was frustrated by her premature discharge from the hospital to move with her parents to another community. Three years later, however, a second diabetic child, complaining of an almost insatiable craving for salt, was subjected to a series of metabolic studies originally designed primarily to determine the natm^e of the physiological disturbance responsible for this bizarre symptom. This patient, a 15-year-old boy, was found to require between 80 and 90 gm. of table salt daily to satisfy his craving. Since he had been on an ordinary diabetic diet (naturally high in K) for a long period of time, had partially depleted his reserve of fixed base as a residt of a mild degree of diabetic acidosis and had vomited an undeter- mined number of times just prior to his admission to the hospital (thereby losing a considerable amotnit of chloride), the inordinate craving for salt was at first presumed to be explainalile on the simple basis of temporary physio- logical need for both Na and CI. That such an explanation was quite inade- cjuate soon became apparent, however, when the craving was observed to 41 8 Relation of Salts to Carbohydrate Metabolism continue long after all evidence of ketosis had disappeared and after a simpli- fied diet with a low K content had been substituted for the ordinary diabetic diet. The possibility that adrenal insufficiency might account for the salt crav- ing was entertained, but no supporting clinical evidence for this complication was obtained. After a preliminary period of observation had demonstrated that the degree of glycosuria in this case, as in the previous one, tended to be inversely pro- portional to the quantity of NaCl ingested, this phase of the problem was investigated more extensively under well-controlled conditions. For the sake of comparison, similar studies were also made on four other juvenile diabetics, who manifested no abnormal taste for salt. Studies were carried out on but two patients at a time. The latter were kept in a small metabolic ward under the constant supervision of special nurses, who collected the urine quantitatively and made certain that meals, salt allowances, and insulin were received by the experimental subjects every six hours precisely as ordered. For the sake of uniformity and accuracy, the basic diet with low Na and K content was made up from powdered whole milk, egg white, egg yolk, unsalted butter, cane sugar, fresh lemon juice (for vitamin C), powdered yeast (for vitamins of the B complex), and water. The mineral constituents of a typical day's diet, containing protein 64, fat 96 and carbo- hydrate 132 gm. were as follows: Na, 1.03; K, 1.34; Ca, 0.68; Mg, 0.16; CI, 1.83; P, 0.93 and S, 0.76 gm. When used, insulin was given at 6-hour intervals in doses sufficiently small to permit some degie of glycosuria in each subdivision of the day, when the patient was on the basic diet. A significant change in the amount of sugar excreted in the mine during an experiment was then regarded as a reliable criterion of an effect produced by the extra salt ingested. The latter was given with or immediately after each of the four meals every day in the form of a 1 per cent solution or in gelatin capsules. Water was given ad libitum. The results of this study may be summarized briefly as follows: Ingestion of excessive amounts (1 to 2 gm. per kilogram of body weight) of NaCl or other sodium salts repeatedly reduced the total output of sugar in the urine of all five patients. As illustrated in figure 1, glycosuria was at times reduced to as little as one-fifth of what it was during control periods. Such an effect was either not apparent or was much less marked, however, when but small amounts of NaCl were ingested or when the ordinary diabetic diet was substituted for the basic experimental diet (low in K). The fasting blood-sugar level was distinctly lower and the respiratory quotient was somewhat higher during periods of excessive Na intake. In the severely diabetic patients, ketosis was found to develop earlier after withdrawal of insulin therapy during periods of ex- tremely low NaCl intake than during periods of excessively high intake. Glycosuria increased rapidly in severe diabetics upon withdrawal of insulin in spite of the high intake of sodium. However, unless the insulin dosage was adjusted to a level to allow hyperglycemia when the low-Na basic diet was Irvine McQuarrie 419 given, administration of 2 gm. of NaCl per kilogram of body weight per day not infrequently resulted in the patient's having typical hypoglycemic reactions. The level of serum K tended to fall and the vninary excretion of K was in- creased slightly by excessive ingestion of NaCl. Both the systolic and the dias- tolic blood-pressure readings were gradually elevated to levels between 30 and 50 per cent above their base lines, as a result of the increased NaCl intake. In the case of one patient, showing the hypercorticoadrenal syndrome with an atypical form of diabetes mellitus, arterial hypertension, and a disturbance Fig. 1. Effects of fiigh-Na, low-K regimen on glycosuria, fasting blood sugar (Bl. Gl.) and respiratory quotient (R.Q.) in severe diabetics. L.R., age 15 years, a salt craver, was more re- sponsive to Na than other diabetics tested. in electrolyte metabolism (increased Na and reduced K in the serum), inges- tion of extra NaCl appeared from a single short test to have a deleterious rather than a beneficial effect (McQuarrie and co-workers"'). In an attempt to learn which, if either, of the two ions, Na or CI, was the more important in producing these changes, KCl was substituted for, or was given in addition to, NaCl in approximately one-third chemically equivalent amounts in a number of separate experiments. As illustrated by the three experiments represented in figure 2, w^hich is largely self-explanatory, the effects of the K salt were found to be diametrically opposite to those produced by Na salts. Glycosuria and fasting blood-sugar levels were gradually increased, while the blood pressure w\is reduced. In a small series of unpublished observa- tions on ordinary juvenile diabetics, Thompson'' subsequently found that the dosage of insulin required for adequate treatment could be reduced signifi- cantly by selecting a diet w^ith a very low K content, other factors remaining unchanged. In an unusually severe case of spontaneous hypoglycemia is an adult patient showing a low concentration of K in the serum, the fasting blood sugar rose from a level around 12 mg. per cent to values varying between 53 420 Relation of Saks to Carbohydrate Metabolism and go mg. per cent and hypoglycemic convulsions ceased, when he was placed on a low-Na high-K regimen (24 to 48 gm. KCl daily) for a few days only. When the KCl was finally refused by the patient because it produced gastric distress, the hypoglycemia and convulsions promptly recurred (McQuarrie and co- workers^"). Attempts by the author and his associates to investigate the influence of varying the oral intake of Na and K on carbohydrate metabolism in depan- creatized and in adrenalectomized dogs did not meet with a high degree of Exper. E, j.P. Insulin ■■ 4unH-s every 6hr. - - t ^ ^ x 0 U( x>s e In ur inej 1 > ^ s ^ J r [ u r » y r \ ^ » y \ f ^ w 1 1 i 10 I"? 20 of cxpcrimc n t No CI intake kCI intake Fig. 2. Antagonistic effects of Na and K on glycosuria and on blood sugar (Bl. Gl.) in dia- betics. Note effectiveness of NaCl in mild diabetic (D.R. exper. D) without insulin. NaCl and KCl Avere fed during different periods in experiments D and E. When given simultaneously with NaCl (exper. F), KCl abolished effects of the Na salt. success because of difficulties involved in making such animals eat the experi- mental diet with the extra salt added or in making them retain the same when given by stomach tube. Ziegler and the author^ studied the effects of Na and K salts in moderate amounts on the glycosuria and hypoglycemia of phloridzin poisoning in four normal adult dogs. No significant effects were demonstrated. In the normal fasting dog, however, continuous administration of KCl was found to impede the development of insulin hypoglycemia (McQuarrie and co-workers^). Since the foregoing studies concerning the antagonistic effects of Na and K on carbohydrate metabolism in diabetes were first reported, a number of publications bearing on some phase of the subject have appeared in the litera- ture. Although a few workers have obtained somewhat equivocal results, most of them have directly or indirectly supported the conclusion that a high-Na low-K regimen increases carbohydrate utilization. McLean^^ and Wilder'" con- firmed the beneficial effects in adult diabetics. The former gave a patient with relative insensitivity to insulin (attributable to infected gangrene of the foot) 40 gm. of NaCl after each of three meals daily. As a result the patient experi- Irvine McQuarrie 421 enced a severe insulin reaction and his rccjuircment lor insulin steadily de- clined thereafter from 115 to 55 units per day while the extra salt was being given. Wilder's experience led him to believe that smaller doses of NaCl were effective in some cases of diabetes. He concluded that the special usefulness of the procedure would be found principally in the management of patients with relative insensitivity to insulin. Insulin requirements were reduced signifi- cantly in a number of his cases so treated. Sanstead^^ reported lowering of the fasting blood sugar in all of eight adult male diabetics to whom he gave from 10 to 90 gm. of NaCl daily. Ingestion of the added salt made it possible to either reduce the insulin dosage or increase the diet of these patients. In a more recent therapeutic trial of the procedure, Strouse and co-workers'^^ reported very variable results when they gave twelve adult diabetics 20 to 30 gm. of NaCl daily supplementary to their ordinary diabetic diets (presumably with high K content). They reported that in any one patient administration of NaCl was followed at one time by decreased glucose excretion and at an- other by increased excretion. One juvenile diabetic given a special diet with fairly low K content for 5 days showed no significant decrease irj glycosuria. Because their patients complained of nausea, however, these authors gave much smaller amounts of NaCl than those given by most of the previous work- ers. Similar variability in results followed administration of KCl. One de- pancreatized dog showed a decrease in glucose excretion from 26.2 gm. (average for 5 control days) to 12.4 gm. per day (average for 7 experimental days), when the Na intake was increased by 3.96 gm. daily. In a second depancreatized dog, showing spontaneous decreases in glucose excretion dtning the course of the experiment, extra NaCl appeared to decrease glycosuria in one experi- ment and to increase it in another. Experiments on a depancreatized goat fed a diet of sucrose and alfalfa, containing 7.8 to 10 gm. of K daily, showed no effect when 2.5 to 3 gm. of Na was given daily. Glycosuria remained fairly marked and the blood sugar was high. This negative effect of NaCl in the presence of so much K in the diet would be expected in view of data presented above. From their variable results these observers concluded that "changes in carbohydrate metabolism which follow ingestion of Na or K chloride cannot be explained as a result of the mere addition of these salts to the dietary intake." Further light has been thrown on the role of NaCl in carbohydrate metab- olism by experimental studies concerned with the effects of deficiency of this salt on carbohydrate tolerance. Michelson^" reported that high blood-sugar values accompanied the NaCl deficiency produced by diuretin. McCance"" later found that five normal human subjects, maintained on NaCl-deficient diets and exposed to profuse sweating, exhibited impaired glucose tolerance. The fasting blood samples, as well as those taken at 30, 60, 90, and 120 minutes after ingestion of 50 gm. of glucose, showed consistently higher glucose values than did normal control subjects. Aldersberg and Wachstein"' observed a pro- found sweeping out of NaCl in experimental animals after complete pan- createctomy and concluded that the pancreas normally controls CI metabolism. 422 Relation o£ Salts to Carbohydrate Metabolism They found that this loss of NaCl intensified the diabetes caused by removal of the pancreas. Glycosuria and the fasting blood sugar were significantly reduced, however, if NaCl was administered either with or without insulin therapy. In several human diabetics tested by these authors, giving of extra NaCl by mouth in moderate amounts (without regard to the K content of the diet) did not appear to modify carbohydrate tolerance so definitely. Calculations based upon carbohydrate intake, decrease in glucose lost in the urine, alterations in blood-sugar level and changes in volttme of body fluids, indicated clearly that the diabetic patients studied by the author and his associates (McQuarrie and co-workers'^") stored glycogen while ingesting large amounts of NaCl. Direct proof that this type of regimen has such an effect, at least in normal animals, was later furnished by the experiments of Crabtree and Longwell."" These authors demonstrated a marked increase in the amount of glycogen in the livers of young rats following ingestion by the latter of ex- cessive quantities of NaCl. Anderson, Herring and Joseph,"^ in a study of the carbohydrate stores in adrenalectomized rats given various amounts of NaCl, found that such rats given 650 to 940 mg. NaCl daily for a period of 176 days after operation were able to store fed glucose almost as well as the normal rat subjected to "sham adrenalectomy." However, animals fed amounts of NaCl distinctly above or below this level were not able to do so. Glycogen stores were lessened in animals given tap water instead of 1 per cent NaCl. Excessive amounts of NaCl (1 234 mg. per day) prevented high glycogen storage. As regards the specific effects of intravenous KCl on the carbohydrate me- tabolism of normal animals, results of various investigators have differed some- what. Semler" early reported finding a sustained hyperglycemia after injection of 3 to 5 gm. of this salt in diabetics. Kiyohara and co-workers'^ obtained similar results from intravenous administration of KCl solution. Kylin and EngeF" reported a slight initial decrease in blood sugar, but did not follow the changes beyond one hour. In the rabbit, Odashima"' found blood sugar to be lowered and lactic acid to be increased. More recently Silvette and Brit- j.Qj^e7,6s reported the production of prolonged hyperglycemia and marked re- duction in liver glycogen in normal rats following injection of KCl solution. Lewis*" found that excessive ingestion of NaCl increased, while KCl de- creased, the sensitivity of normal rats to insulin. Animals with a high KCl intake stored less glycogen following dextrose administration than control rats. Those with an excessive intake of NaCl stored more glycogen but oxidized less than control animals. Fasting blood-sugar levels and the carbohydrate tolerance of normal rats were not found to be altered significantly by these salts. In partially depancreatized, diabetic rats and in rats of the Yale strain with hereditary diabetes, Orten and Devlin'"' '"demonstrated that carbohydrate tolerance could be restored to essentially normal levels by the administration of NaCl in large amounts without insulin. Sayers and Orten" found an in- creased storage of glycogen in the livers of such diabetic rats after periods of high NaCl intake. Irvine McQuairie 423 Serum K is characteristically elevated in severe, inadequately controlled diabetes mellitus. Rathery and Bertoliatti''^ found K values to average 35 per cent above normal in seven cases of diabetes not on insulin therapy and 22 per cent above normal in seven insulin-treated cases. In a comprehensive study of the serum electrolytes and acid-base equilibrium in four cases of severe diabetes, Guest'^ found the Na and CI to be abnormally low and the K and inorganic P to be gieatly elevated, when the patients were out of control. 7 o him these changes appeared to be more closely related to prolonged duration of the accompanying acidosis and to temporary impairment of renal function than to the intensity of the acidosis, as indicated by the pH and the CO, con- tent of the plasma. Insulin therapy caused a specific reversal of all of these alterations, the serum K and inorganic P declining to subnormal levels. Comment It is evident, from the miscellaneous data presented here, that sodium and potassium play important roles in carbohydrate metabolism. While there is obviously a lack of complete agreement between the results of various workers concerned with the problem of such interrelationships, differences in experi- mental conditions apparently account for most of the discrepancies reported. The mechanisms by which the Na and K ions influence carbohydrate economy and those by which their exchanges in the body are in turn influenced by it are not known. The roles of the two elements are obviously different and are in the main antagonistic to each other. Potassium would appear to have the better claim to an essential role in the intermediary metabolism. Its movements into and out of the blood in company with glucose and inorganic PO^ or with lactic acid under various conditions, have suggested that it may be organically bound with some hexosephosphate compound or may be adsorbed to the surface of the latter. The fact that the quantity of K involved in the deposition of glyco- gen or released with its breakdown is far less than the glycogen (measured in equivalents), argues against the assumption that it enters into direct chem- ical combination with the carbohydrate as PO^ does in the process of phos- phorylation. At present it would appear more likely that K plays the role of accelerating certain enzymatic reactions involved in carbohydrate metabo- lism, just as magnesium, manganese and cobalt have been sho^vn by Cori'^ to do in the case of the enzyme, phosphoglucomutase. It has already been shown to accelerate the fermentative and oxidative breakdown of glucose in yeast cultures and in fresh tissue slices. The suggestion that its physical relationship to the complex carbohydrate molecule is one of adsorption becomes more likely in the light of this latter role. Sodium does not move into or out of the blood or other media with glucose and inorganic PO^ (Ziegler™). Nor is it known to activate enzymatic reactions under simple experimental conditions similar to those in which the accelerat- ing effect of K has been demonstrated. Its influence on carbohydrate metabo- 424 Relation o£ Salts to Carbohydrate Metabolism lism appears to be an "over-all" action in the animal body. While the possi- bility of its acting directly on certain enzyme systems has not been entirely excluded, its effects appear at present to be related to endocrine or autonomic nervous functions. When administered in large amounts, Na salts have been shown to increase the deposition of glycogen in the liver of normal, diabetic and adrenalectomized (glucose-fed) rats. There is some evidence to indicate that the rate of oxidation of glucose is simultaneously diminished. In these respects NaCl behaves very much like certain adrenal cortical hormones. Long and his co-workers" have demonstrated that the latter slow up glucose oxida- tion while accelerating the deposition of glycogen in the liver. Since this glycogen-conserving action of Na salts is effective even in totally adrenalec- tomized animals, when they are supplied with glucose, it is obvious that the Na effect is not dependent entirely upon the adrenal gland. Recent experi- ments by Grollman,^* however, emphasize the interdependence between the function of the adrenal cortex and the action of sodium salts. The results of studies by Richter™'^ on the appetites of adrenalectomized rats for NaCl and dextrose likewise indicate the essential interdependence of Na and carbohy- drate metabolism in relationship to adrenal function. Data reported by various workers from experiments on normal and diabetic animals, as well as on diabetic human subjects, indicate that Na increases sensitivity to administered insulin. In the intact animal, it may act directly on the insulin-producing mechanism or, on the other hand, it may serve to inhibit the so-called "anti-insulin" or "diabetogenic" mechanisms. Whether or not the production of glycogen is increased at the expense of protein, when Na salts are given in excessive amounts, has not as yet been ascertained; nor has the significance of differences in the effects of Na and K on water balance been fully determined. In addition to its supposed roles in the intermediary metabolism of carbo- hydrates, K also appears to act on the intact animal in an "over-all" way with effects antagonistic to those of Na. When administered in comparatively large amounts to normal or diabetic subjects, it inhibits the formation of glycogen or actively promotes glycogenolysis with resulting hyperglycemia and glyco- suria (in diabetics). Whether or not these effects are mediated through neuro- endocrine mechanisms or are due to direct influence on the combined action of the phosphorylase and phosphatase systems, considered by Cori^^ to govern the formation of blood sugar from liver glycogen, cannot at present be said. Summary Miscellaneous data pertaining to the interrelationships between the exchanges of sodium and potassium and carbohydrate metabolism have been assembled from widely scattered sources. A critical evaluation of the fairly imposing array of evidence thereby made available indicates that these interrelations are of profound significance in both animal and plant economy. The "over-all" effects of these two monovalent bases in the animal body have Irvine McQiiarrie 425 been shown to be antagonistic to each other in the main. Sodium tends to re- tard the oxidation of glucose and to increase the deposition of glycogen in the liver. On the other hand, potassium inhibits glycogenesis in the liver and tends to increase blood sugar and glycosuria at the expense of liver glycogen. Under certain conditions it apparently activates the enzymatic breakdown of glucose. The clinical significance of these interrelationships is apparent in such con- ditions as diabetes mellitus, hyperinsulinism, Addison's disease and familial periodic paralysis. REFERENCES i.Murlin, }. R.. and Kramer, B.: Jl. Biol. Chem. 27:481, 517, 1916. 2. Murlin, J. R., and Graver, J. A.: Jl. Biol. Chem. 28:289, 1916. 3. Underbill, F. P.: |1. Amer. Med. Assn. 68:497, 1917. 4. Beard, A. H.: Archs. Internal Med. 21:716, 1918. 5. Thompson, G. D.; Mitchell, M., and Kolb, L. C.: Biochem. Jl. 27:1253, 1933. 6. Johnston, J. A., and Maroney, J. W.: Amer. Jl. Diseases Childr. 49: 1240, 1935. 7.Schenk, H.: Ztschr. f. d. gsmt. exper. Med. 12:269, 1921. 8. Abelin, J., and Goldener, E.: Klin. Wchnschr. 4:2466, 1925. 9. Semler, R.: Klin. Wchnschr. 4:697, 1925. 10. Glass, E., and Beiless, I.: Ztschr. f. d. gsmt. exper. Med. 73:801, 1930. 1 1. Briggs, A. P.; Koechig, I.; Doisy, E. A., and Weber, C. J.: Jl. Biol. Chem. 58:721, 1923. 12. Harrop, G. A., and Benedict, E. M.: Jl. Biol. Chem. 59:683, 1924. 13. Wigglesworth, V. E.; Woodrow, C. E.: Smith. \V., and Winter, L. B.: Jl. Physiol. 57:447, 1922. 14. Hausler, H.. and Heesch, O.: Pfliigers Arch. i. d. gsmt. Physiol. 210:545, 1925. 15. Kerr, S. E.: Jl. Biol. Chem. 78:35, 1928. 16. Cori, C. F., and Cori, G. T.: Naunyn-Schmiedebergs Arch. f. exper. Pathol, u. Pharmakol. 172:249, 1933. 17. Ellsworth, R., and Weinstein, A.: Bull. Johns Hopkins Hosp. 53:21, 1933. 18. Keys, A.: Amer. Jl. Physiol. 123:608, 1938. 19. McQuarrie, I., and Ziegler, M. R.: Proc. Soc. Exper. Biol. & Med. 39:523, 1938. 20. Pulver, R., and Verzar, F.: Nature (London) 145:823, 1940. 21. Day, D., and Comboni, S.: Amer. Jl. Bot. 24:594, 1937. 22. Lasnitski, A., and Sziirenyi, E.: Biochem. Jl. 29:580, 1935. 23. Lasnitski, A.: Biochem. Ztschr. 264:285, 1933. 24. Lasnitski, A.: Protoplasma 22:274, 1934- 25. Ashford, C. A., and Dixon, K. C: Biochem. Jl. 29: 157, 1935. 26. Dickens, F., and Greville, G. D.: Biochem. Jl. 29:1468, 1935. 27. Fenn, W. O.: Jl. Biol. Chem. 128:297, 1939. 28. Flock, E.; Bollman, J.L.; Mann, F. C, and Kendall, E.G.: Jl. Biol. Chem. 125:57, 1938. 29. Kendall, E. C; Flock, E. V.; Bollman. J. L., and Mann, F. C: Jl. Biol. Chem. 126:697, 1938. 30. Long, C. N. H.; Fry, E. G., and Thompson, K. W.: Amer. Jl. Physiol. 123:130, 1938. 31. Lasnitzki, A.: Xatiire (London) 146:99, 1940. 32. Verzar, F., and Somogyi, J. C: Nature (London) 144: 1014, 1939. 33. Verzar, F., and Somogyi. J. C: Helvet. med. act. 23: 1087, 1940. 34. Verzar, F., and Somogyi, J. C: Archs. internatn. de pharmacoiUn. vi de ihcrapic 65:17, 1941. 35. Fenn, W. O.: Physiol. Revs. 16:450, 1936. 36. DSilva, J. L.: Jl. Physiol. 82:393, 1934. 37. DSilva, J. L.: Jl. Physiol. 90:303, 1937. 38. Marenzi. A. D., and Gerschman, R.: Rev. Soc. aigenlin. de hiol. 12: |2 |. 1936. 426 Relation of Salts to Carbohydrate Metabolism 39. Marenzi, A. D., and Gerschman, R.: Cpts. rd. Soc. de biol. 124:382, 1937. 40. Schwartz, H.: Naunyn-Schmiedcbergs Arch. f. expcr. Pathol, u. Pharmakol. 177:628, 1935. 41. Houssay, B. A.; Marenzi, A. D., and Gerschman, R.: Cpts. rd. Soc. de biol. 124:383, 1937. 42. Snyder, C. D., and Johnson, R. E.: Johns Hopkins Hosp. Bull. 62:1 10, 1938. 43. Fenn, W. O.: Physiol. Revs. 20:377, 1940. 44. Soskin, S.: Amer. Jl. Physiol. 81:382, 1927. 45. Cori, C. F.: Physiol. Revs. 11:143, 1931- 46. Walker, M.: Lancet 229:47, 1935. 47. Aitken, R. S.; Allott, E. N.; Castleden, L. I. M., and Walker, M.: Clin. Sci. 3:47, 1937. 48. Allott, E. N., and McArdle, B.: Clin. Sci. 3:229, 1938. 49. McQuarrie, I., and Thompson, W. H.: Proc. Soc. Exper. Biol. & Med. 31:907, 1934. 50. McQuarrie, I.; Thompson, W. H., and Anderson, J. A.: Jl. Nutrit. 1 1 :77, 1936. 51. McQuarrie, I.; Ziegler, M. R., and Johnson, R. M.: Endocrinology 21:762, 1937. 52. Thompson, W. H.: Unpublished data. 53. Ziegler, M. R., and McQuarrie, I.: Proc. Soc. Exper. Biol. & Med. 39: 142, 1938. 54. McQuarrie, I.; Ziegler, M. R., and Stone, W.E.: Chin. Med. Jl. 58:1, 1940. 55. McLean, A. R.: Proc. Staff Meetgs. Mayo Clin. 10:321, 1935. 56. Wilder, R. M.: Archs. Internal Med. 57:422, 1936. 57. Sanstead, H. R.: Hosp. News 21:1, 1936. 58. Strouse, S.; Buel, F.; Kay, R., and Drury, D.: Jl. Nutrit. 21:599, 1941. 59. Michelson, H.: Naunyn-Schmiedebergs Arch. f. exper. Pathol, u. Pharmakol. 173:750, 1933. 60. McCance, K. A.: Biochem. Jl. 31 :i276, 1937. 61. Aldersberg, D., and Wachstein, M.: Klin. Wchnschr. 16:85, i937- 62. Crabtree, D. G., and Longwell, B. B.: Proc. Soc. Exper. Biol. & Med. 34:705, 1936. 63. Anderson, E.; Herring, V., and Joseph, M.: Proc. Soc. Exper. Biol. & Med. 45:488, 1940. 64. Kiyohara, K.; Morita, M., and Muta, S.: Cpts. rd. Soc. de biol. 120:1011, 1935. 65. Kylin, E., and Engel, A.: Klin. Wchnschr. 4:653, 1925. 66. Odashima, G.: Tohoku Jl. Exper. Med. 18:250, 1932. 67. Silvette, H., and Britton, S. W.: Proc. Soc. Exper. Biol. & Med. 37:252, 1937. 68. Silvette, H.; Britton, S. W., and Kline, A.: Amer. Jl. Physiol. 122:446, 524, 1938. 69. Lewis, R. C: Univ. Colorado Studies, genrl. s. (A) 26(3):73, 1940. 70. Orten, J. M., and Devlin, H. B.: Proc. Soc. Exper. Biol. &: Med. 42:632, 1939. 71. Orten, J. M., and Devlin, H. B.: Jl. Biol. Chem. 136:461, 1940. 72. Sayers, G., and Orten, J. M.: Proc. Soc. Exper. Biol. & Med. 46:287, 1941. 73. Rathery, F., and Bertoliatti, J.: Cpts. rd. Soc. de biol. 116:1346, 1934. 74. Guest, G. M.: Personal communication. 75. Cori, C. F.: Endocrinology 26:285, 1940. 76. Ziegler, M. R.: Proc. Soc. Exper. Biol. & Med. 43: 165, 1940. 77. Long, C. N. H.; Katzin, B., and Fry, E. G.: Endocrinology 26:309, 1940. 78. Grollman, A.: Endocrinology 29:862, 1941. 79. Richter, C. P., and Eckert, J. F.: Endocrinology 22:214, 1938. 80. Richter, C. P.: Endocrinology 29:111, 1941. HARVEY'S IDEAS OF EMBRYONIC NUTRITION By A. W. MEYER From the DEPARTMENT OF ANATOMY STANFORD UNIVERSITY, CALIFORNIA HARVEY'S IDEAS OE EMBRYONIC NUTRITION* IN HIS LETTER of April 28, 1652, to R. MoiTson, M.D., of Paris, Harvey regarded it as "... a most certain fact . . . that the embryos of all red-blooded animals are nourished by means of the umbilical vessels from the mother, and this in virtue of the circulation of the blood. They are not nourished, however, im- mediately by the blood, as many have imagijied, but after the mariner of the chick in ovo, which is first nourished by the albumen, and then by the vitellus . . ." (p. 608). After considering conception in the deer in the month of December, Harvey had concluded that: "From all of what precedes it is manifest that in both the classes of viviparous animals alluded to, those, namely, that are provided with carunculae or cotyledons, and those that want them, and perhaps in viviparous animals generally, the foetus in utero is not nourished otherwise than the chick in ovo; the nutritive matter, the albumen, being of the same identical kind in all. . . . And this is further obvious from the fact of the extremities of the umbilical vessels, when they are drawn out of the afore-mentioned mucor, looking completely white; a certain proof that they absorb this mucilage liquefied only, and not blood. The same arrangement may very readily be observed to obtain in the egg. "The human placenta is rendered uneven on its convex surface, and where it adheres to the uterus, by a nimiber of tuberous projections, and it seems indeed to adhere to the uterus by means of these; it is not consequently at- tached at every point, but at those places only where the vessels pierce it in search of nourishment, and at those where, in consequence of this arrange- ment, an appearance as if of vessels broken short off is perceived. But whether the extremities of these vessels suck up blood from the uterus, or rather a cer- tain concocted matter of the nature of albuynen, as I have described the thing in the hind and doe, I have not yet ascertaiiied" (p. 497). "It seems manifest, therefore, that the foetus in utero is not nourished by its mother's blood, but by this albuminous fluid duly elaborated. It may even be, perhaps, that the adult animal is not nourished immediately by the blood, but rather by something mixed with the blood, which serves as the ultimate ali- ment; as may perhaps be more particularly shown in our Physiology^ and particular treatise on the Blood" (p. 498). In his letter of July 13, 1655 (old style), to "John Dan. Horst," he likewise declared, "I only say (keeping silence as to any other channels), that the nutri- * All quotations not separately identified are from \Villis' cdiiion (London: 1847) of Harvey's works. The italics and words in brackets were introduced by the compiler. More annotations and comments were precluded bv the unavoidable limits set. f Harvey is not known to have written such a treatise. [4293 430 Harvey's Ideas o£ Embryonic Nutrition tive juice might be as readily transported by the uterine arteries, and distilled into the uterus, as watery fluid is carried -by the emulgent arteries to the kid- neys . . . What you say of the excrements of new-born infants differing from those of the child that has once tasted milk I do not admit . . ." (p. 614-615). Harvey began his discussion "of the nutrition of the chick in ovo," in "Ex- ercise the Fifty-eighth" of the De Generatione, by recalling that Democritus, Epicurus, and Hippocrates supported the idea that the mammalian foetus sucks in utero. He said that his teacher, Fabricius, had rejected this idea "as a delusion," but that he nevertheless believed it because his observations on generation make that opinion not "merely probable" but necessary. In support of his opinion Harvey further called attention to the "fact" that the amniotic fluid ". . . has a pleasant taste, like that of watery milk, so that almost all viviparous animals lap it up, . . . greedily swallowing" it (p. 435).* "To me, therefore, the opinion of Hippocrates appears more probable than that of Fabricius and other anatomists, who look on this liquid as sweat, and believe that it must prove detrimental to the foetus. I am disposed, I say, to believe that the fluid with which the foetus is surrounded may serve it for nourish- ment; that the thinner and purer portions of it, taken up by the umbilical veins, may serve for the constitution and increase of the first formed parts of the embryo; and that from the remainder or the milk,-j- taken into the mouth by suction, passed on to the stomach by the act of deglutition, and there digested or chylified, and finally absorbed by the mesenteric veins, the new being continues to grow and be nourished. I am the more disposed to take this view from certain not impertinent arguments, which I shall proceed to state. "As soon as the embryo acquires a certain degiee of perfection it moves its extremities, and begins to prove the actions of the organs destined to locomo- tion. Now I have seen the chick in ovo, surrounded with liquid, opening its njouth, and any fluid that thus gained access to the fauces must needs have been swallowed; for it is certain that whatever passes the root of the tongue and gains the top of the oesophagus, cannot be rejected by any animal with a less effort than that of vomiting. ... If the embryo swimming in the fluid in question, then, do but open his mouth, it is absolutely necessary that the fluid must reach the fauces; and if the creature then move other muscles, wherefore should we not believe that he also uses his throat in its appropriate office and swallows the fluid? * Even as late as 1815, Beclard, on the basis of experimental evidence, concluded that the fetuses of mammals swallow amniotic fluid because hairs were present in the meconium of fetuses whose necks had been constricted by a tourniquet. Beclard also had found hairs in a still born proximal to the site of an intestinal obstruction, and had noticed that the increase in the rapidity of fetal respiratory movements was proportional to the extent of placental detachment. Moreover, pigment injected into the amniotic fluid, after ligation of the fetal neck, was found up to the ligature. fThe word milk was then used to designate the fluids surrounding the fetus instead of those within the uterus, but Harvey also spoke of milk in the thymus and lymph glands. A. W. Meyer 43 1 "It is further quite certain that in the crop of the chick,— and the same thing occurs in reference to the stomach of other embryos— there is a certain matter having a colour, taste, and consistence, very similar to that of the liquid men- tioned, and some of it in the stomach digested to a certain extent, like coagu- lated milk; and further, whilst we discover a kind of chyle in the upper intestines, we find the lower bowels full of stercoraceous excrements. In like manner we perceive the large intestines of the foetuses of viviparous animals to contain excrements of the same description as those that distend them when they feed on milk. In the sheep and other bisulcated animals we even find scybala" (pp. 438-439). This he said was observed by "Volcher Goiter, a careful and experienced dissector." Harvey concluded that "The embryo, therefore seeks for and sucks in nourishment by the mouth; and you will readily be- lieve that he does so if you rip him from his mother's womb and instantly put a finger in his mouth; which Hippocrates thinks he would not seize had he not previously sucked whilst in the womb" (p. 440). "In the stomach of the foetus there is a watery fluid contained, not unlike that in which it swims, but somewhat more turbid or less transparent. It re- sembles the milk that begins to be secreted in the breasts of pregnant women about the fourth or fifth month of pregnancy, and may be pressed out of the nipples, or it is like the drink which we call white posset. "In the small intestines there is an abundance of chyle concocted from the same matter; in the colon greenish faeces and scybala begin to appear" (pp. 493-494)- "Just as the colliquament found in the crop of the chick is a kind of crude milk, whilst the same fluid discovered in the stomach is concocted, white, and curdled; so in viviparous animals, before the milk is concocted in the mammae, a kind of dew and colliquament makes its appearance there, and the colli- quament only puts on the semblance of milk after it has undergone concoction in the stomach. And so it happens, in Aristotle's opinion, that the first and most essential parts are formed out of the purer and thinner portion of the colliquament, and are increased by the remaining more indifferent portion after it has undergone elaboration by a new digestion in the stomach. In the same way are the other less important parts developed and maintained. Thus has nature, like a fond and indulgent mother, been sedulous rather to provide superfluity, than to suffer any scarcity of things necessary. Or it might be said to be in conformity with reason to suppose that the foetus, noAV gro-^vn more perfect, should also be nourished in a more perfect manner, by the mouth, to wit, and be a more perfect kind of aliment, rendered purer by having under- gone the two antecedent digestions and been thereby freed from the two kinds of excrementitious matter. In the beginning and early stages, nourished by the ramifications of the umbilical veins, it leads in some sort the life of a plant; the body is then crude, white, and imperfect; like plants, too, it is motionless and impassive. As soon, however, as it begins by the mouth to partake of the 432 Harvey's Ideas o£ Embryonic Nutrition same aliment farther elaborated, as if feeling a diviner influence, boasting a higher grade of vegetative existence, the gelatinous mass of the body is changed into flesh, the organs of motion are distinguished, the spirits are perfected, and motion begins; nor is it any longer nourished like a vegetable, by the roots, but, living the life of an animal, it is supported by the mouth" (p. 441-442). "These and other observations of the same kind [upon prenatal deglutition, suckling, and intestinal contents] make it extremely probable that the chick in ovo is nourished in a twofold manner, namely, by the umbilical and by the mesenteric veins. By the former he imbibes a nourishment that is well nigh perfectly prepared, whence the first-formed parts are engendered and aug- mented; by the latter he receives chyle for the structure and growth of the other remaining parts" (p. 440). "All admit this distinction of fluids. But I, as I have already said, distinguish two albumens in the egg, kept separate by an interposed membrane, the more external of which embraces the other within it, in the same way as the yolk is surrounded by the albumen in general. I have also insisted on the diverse nature of these albumens; distinguished both by situation and their surround- ing membranes, they seem in like manner calculated for different uses. Both, however, are there for ends of nutrition, the outermost, as that to which the branches of the umbilical veins are earliest distributed, being first consumed, and then the inner and thicker portion; last of all the vitellus is attacked, and by it is the chick nourished, not only till it escapes from the shell but for some time afterwards. "But upon this point we shall have more to say below, when we come to speak of the manner in which the foetuses of viviparous animals are developed, and at the same time demonstrate that these all derive their origin from eggs, and live by a twofold albuminous food in the womb. One of these is thin- ner, and contained within the ovum or conception; the other is obtained by the umbilical vessels from the placenta and uterine cotyledons. The fluid of the ovum resembles a dilute albumen in colour and consistence; it is a sluggish, pellucid liquid, in all respects similar to that which we have called the col- liquament of the egg, in which the embryo swims, and on which it feeds by the mouth. The fluid which the foetus obtains from the uterine placenta by the aid of the umbilical vessels is more dense and mucaginous, like the inspissated albumen. Whence it clearly appears that the foetus in utero is no more nour- ished by its parent's blood than is the suckling afterivards, or the chick in ovo; but that it is nourished by an albuminous matter concocted in the placenta, and not ujilike white of egg" (p. 447-448). The placenta "in like manner [as the liver], prepares for the foetus alimen- tary matters which have come from the mother . . ." and ". . . so, too, the placenta . . . abounds in an albuminous fluid, and is only to be found at the period of pregnancy" (p. 563). Harvey did not hold these views because he rightly believed that the blood A. W. Meyer 433 of the mother does not circulate in the fetus, but because he held that in pre- natal life the placenta plays the nutritional role that the mother plays in post- natal life, and when discussing the role of the "humours"*— or uterine and embryonic fluids— he declared, "... I nevertheless most confidently assert my belief that these humours are at the commencement destined for the nourish- ment of the foetus, just as the colliquament and albumen are in the case of the chick; but that, in course of time, when the thinner and purer portions are absorbed, the remainder takes on the character of excrementitious matter, but still has its uses, and in some animals especially conduces to the safety of the foetus, and also greatly facilitates birth. For just as wine becomes poor and tasteless when the spirit has evaporated; and as all excreted matters owe their origin for the most part to what has been previously food; so, after all the nutrient portions of the fluid contained in the chorion have been taken up by the foetus, the remainder becomes excrementitious, and is applied to the above-mentioned uses. But all the fluid of the amnioji is usually consumed by the time of birth; so that it is probable the foetus seeks its exit on account of deficiency of nutriment" (p. 559). Although he thought of the conceptus as becoming attached to the uterus for the purpose of taking up nutriment, Harvey emphasized that "At the commencement the 'conception' (like an egg placed within the uterus) is found in contact with every part of the uterus, yet at no point is it adherent; but produces and nourishes the embryo out of the humours contained within it, as I have explained in the instance of the hen's egg. This adhesion, or giowing together, first takes place, and the fleshy mass (constituting the bond of union between the 'conception' and the uterus) is first produced, when the foetus becomes perfectly formed, and, through want either of different or more abun- dant notirishment, dispatches the extremities of the umbilical vessels to the uterus, that from hence (as plants do from the earth by their radicles), it may absorb the nutrient juices. For in the beginning, as I have said, when the 'punctum saliens' and the blood can alone be seen, the ramifications of the umbilical vessels are only visible in the colliquament and amnion. When, however, the fabric of the body is completely formed, the ramifications extend further, and are distributed in vast numbers throughout the chorion, that from the albuminous fluid which there exists, they may obtain nourishment for the foetus. "Hence it is manifest that the young of viviparous animals are at the begin- ning nourished in exactly the same manner as the chick in the eg^; and that they are detained within the uterus in order that (when they can no longer supply themselves with nutriment from their own stores) they may form ad- hesions to it by means of this fleshy substance, and receiving more abundant supplies of food from the mother, may be nourished and made to grow. * By humours, as here used, Harvey understood both uterine and fetal fluids but he also thought of the ovum or primordium of the new being, as "formed out of the humours of the lUerus. " 434 Harvey's Ideas of Embryonic Nutrition "Wherefore Fabricius has rightly observed, that in some animals the 'con- ception' is scarcely attached to the uterus at all. Thus the sow and the mare have no such fleshy mode of union,— but in them the ovum or 'conception,' as in the beginning it is formed out of the humours of the uterus, so it is nourished subsequently by the same means; just as the ovum of the hen is supplied with aliment at the expense of the albuminous matter without any connexion whatever with the uterus: and thus the foetus is furnished with aliment by the 'conception' in which it is contained, and is nourished as the chick is from the fluids of the egg. This is a strong argument that the foetus of viviparous animals is no more nourished by the blood of the mother than the chick in the egg; and moreover, that the fluid within the chorion is neither urine nor any other excrementitious matter; but serves for the support of the foetus. Although as I have before remarked, it is possible when all the nutrient portions have been taken up, the remainder may degenerate into excrementi- tious matter resembling urine. This is also clear from what I formerly observed of the cotyledons in the deer, viz., that in these animals the fleshy mass was of a spongy character, and constituted, like a honeycomb, of innumerable shallow pits filled with a muco-albuminous fluid (a circumstance already ob- served by Galen); and that from this source the ramifications of the umbilical vessels absorbed the nutriment and carried it to the foetus: just as, in animals after their birth, the extremities of the mesenteric vessels are spread over the coats of the intestines and thence take up chyle" (p. 561-562). In regard to the role of the so-called "spirits" in nutrition, he asserted that "They who advocate incorporeal spirits have no ground of experience to stand upon; their spirits indeed are synonymous with powers or faculties, such as a concoctive spirit, a chylopoietic spirit, a procreative spirit, etc.— they admit as many spirits, in short, as there are faculties or organs" (p. 116). "For if the spirits exhaling from the blood, like the vapour of water at- tenuated by heat, exist in a state of constant flow and succession as the pabu- lum of the tissues, it necessarily follows that they are not distinct from this pabulum, but are incessantly disappearing; whereby it seems that they can neither have influx nor reflux, nor passage, nor yet remain at rest without the influx, the reflux, the passage (or stasis) of the blood, which is the fluid that serves as their vehicle or pabulum" (p. 118). He remained skeptical regarding the role of the lacteals and lymphatics in nutrition for he held that "Even as the umbilical veins absorb the nutritive juices from the fluids of the egg and transport them for the nutrition and growth of the chick, in its embryo state, so do the meseraic veins suck up the chyle from the intestines and transfer it to the liver; and why should we not maintain that they perform the same office in the adult?" (p. 95). In his first letter to Riolan, written in 1649, Harvey further declared, "I shall elsewhere state my views of the lacteal veins when I treat of the milk found in different parts of new-born animals, especially of the human subject; for it is met with A. W. Meyer 435 in the mesentery and all its glands, in the thymus, in the axillae, also in the breasts of infants" (p. 95). He spoke of this also in his letter of April 28, 1 652, to "R. Morison, M.D., of Paris," in which he wrote: "But for various reasons, and led by several experiments, I could never be brought to believe that that milky fluid [in the lacteals] was chyle conducted hither from the intestines, and distributed to all parts of the body for their nourishment; but that it was rather met with occasionally and by accident, and proceeded from too ample a supply of nourishment and a peculiar vigour of concoction; in virtue of the same law of nature, in short, as that by which fat, marrow, semen, hair, etc., are produced; even as in the due digestion of ulcers pus is formed, which the nearer it approaches to the consistency of milk, viz., as it is -^vhiter, smoother, and more homogeneous, is held more laudable, so that some of the ancients thought pus and milk were of the same nature, or nearly allied. Wherefore, although there can be no question of the existence of the vessels themselves, still I can by no means agree with Aselli in consider- ing them as chyliferous vessels, and this especially for the reasons about to be given, which lead me to a different conclusion. For the fluid contained in the lacteal veins appears to me to be pure milk, such as is found in the lacteal veins [the milk ducts] of the mammae. Now it does not seem to me very probable (and more than it does to Auzotius in his letter to Pecquet) that the milk is chyle, and thus that the whole body is nourished by means of milk. The reasons ^vhich lead to a contrary conclusion, viz., that it is chyle, are not of such force as to compel my assent. I should first desire to ha\e it demonstrated to me by the clearest reasonings, and the guarantee of experiments, that the fluid contained in these vessels was chyle, which, brought hither from the intestines, supplies nourishment to the whole body. For unless we are agreed upon the first point, any ulterior, any more operose, discussion of their nature, is in vain. But how can these vessels serve as conduits for the whole of the chyle, or the nourishment of the bodv, when we see that thev are different in different animals? In some they proceed to the liver, in others to the porta only, and in others still to neither of these. In some creatures they are seen to be extremely numerous in the pancreas; in others the thymus is crowded with them; in a third class, again, nothing can be seen of them in either of these organs. In some animals, indeed, such chyliferous canals are nowhere to be discovered (vide Liceti Epist. xiii, lit. ii, p. 83, et Sennerti Praxeos, lib. v, tit. 2, par. 3, cap. 1); neither do they exist in any at all times. But the vessels which serve for nutrition must necessarily both exist in all animals, and present themselves at all times; inasmuch as the waste incurred by the ceaseless efflux of the spirits, and the wear and tear of the parts of the body, can only be supplied by as ceaseless a restoration or niurition. And then, their very slender calibre seems to render them not less inadequate to this duty than their structure seems to unfit them for its performance: the smaller channels ought plainly to end in larger ones, these in their turn in channels larger still, and the whole to concentrate in one great trunk, which 43^ Harvey's Ideas of Embryonic Nutrition should correspond in its dimensions to the aggregate capacity of all the branches; just such an arrangement as may be seen to exist in the vena portae and its tributaries, and farther in the trunk of the tree, which is equal to its roots. Wherefore, if the efferent canals of a fluid must be equal in dimensions to the afferent canals of the same fluid, the chyliferous ducts which Pecquet discovers in the thorax, ought at least to equal the two ureters in dimensions; otherwise they who drink a gallon or more of one of the acidulous waters could not pass off all this fluid in so short a space of time by these vessels into the bladder. And truly, when we see the matter of the urine passing thus copiously through the appropriate channels, I do not see how these veins could preserve their milky colour, and the urine all the while remain without a tinge of whiteness" (pp. 605-606). Almost wholly unencumbered by personal comments, the above words of one of the immortals of science are offered in cordial tribute to a valued friend and fruitful seeker who not only has garnered and treasured many precious volumes from seekers of days long passed but has generously entrusted them to others.— May Time deal gently with him and ease the burden of passing years. OBSERVATIONS ON THE PATHOGENESIS OF UNDULANT FEVER By K. F. MEYER From the GEORGE WILLIAMS HOOPER FOUNDATION UNIVERSITY OF CALIFORNIA SAN FRANCISCO, CALIFORNIA OBSERVATIONS ON THE PATHOGENESIS OF UNDULANT FEVER CONTINUOUS perusal of the many papers which have appeared during the past fifteen years on human brucellosis, induced by the abortus and siiis types of Brucella, makes it appear that this disease still remains an inade- quately understood and consequently an intriguing subject. Despite the vast amount of research and the collection of case histories, relatively little is known concerning the genesis and pathology of the infection. In fact, what has been reported is frequently contradictory and little agreement has as yet been reached. In order to elucidate certain phases of the morbid processes, the pathological and bacteriological findings made on a laboratory-animal care- taker, who succumbed to a fulminating Br. suis septicemia within eleven days after he was observed to be ill, are herewith reported. The history of the patient presents the following essential data: White male, D.S., age 24, laboratory technician with a strongly positive phagocytic index test of 6.88 for Br. abortus and a positive allergic skin test indicative of a passed latent Brucella infection, complained on November 24 of discomfort in the abdomen, nausea, vomiting after a week of malaise, marked spells of perspiration, and occipital headache. The examination disclosed a purplish hue of the skin over his back, face and anterior thorax, a temperature of 101° F., tachycardia (128), vesicular breathing, a soft large spleen; he was decidedly restless and irritable, occasionally irrational. R.B.C. 4,640,000; W.B.C. 13,200; 91 per cent P.M.N, and 4 per cent monocytes; urine: albumen-1-H-, small hyaline and granular casts. A quantitative blood culture yielded 1,750 colonies of Br. suis per cc. On November 27, a slight icteric tinge of the sclera was noted and the urine ^vas positive for urobilinogen. He developed a progressive jaundice (icterus index 130. van den Bergh direct and indirect); the serum gave a positive agglutination test with Br. abortus 1 :4o. On December 2, extensive consolidation at the base of the right lung was recorded; he became comatose and expired on December 4, despite repeated transfusion and glucose administrattion by clysis and supportive treatment. Autopsy: The autopsy was performed 7 hours after death by W. T. Partch, M.D., and the author. Only the significant findings as dictated are recorded. The body of a ^veil-nourished male about 30 years of age shows marked icteric tinge to the skin and post-mortem lividity. On incision in the abdominal cavity and in the right pleural cavity approximately 50 cc. of slightly blood-tinged fluid are found. No exudate is seen in the left cavity while the pericardial cavity contains 50 cc. of clear straw-colored fluid. The weight of the heart is 370 gm. The valves and the myocardium are essentially normal. Left Lung: A slight amount of fibrinous exudate is present over the visceral pleura of the left lower lobe. In cut section the entire lobe is seen to be markedly reddened and edematous (hemorrhagic edema). No consolidation is present; mediastinal lymph nodes are reddened, soft, and markedly enlarged. Riglit Lung: On palpation of the upper lobe and the posterior aspect of the right lower lobe many areas of consolidation are felt; the middle lobe crepitates normally throughout. The upper lobe on section is red in appearance, and practically consolidated. There is central consolidation in the base. From numerous nodules along the posterior asfject of the lower lobe droplets of pus exude from the cut surface. There is a very recent fibrinous exudate over the visceral pleura of the right lung. The Spleen ^veighs 420 gm.; it is moderately firm on palpation and dark bluish in color. 1:439] 440 Pathogenesis o£ Undulant Fever On section the Malpighian bodies are not always visible and the splenic substance is quite pulpy in character. Tlie Intestines show a slight superficial injection throughout the entire tract. The mesen- teric blands are not enlarged. The mucosa of the terminal end of the small intestine shows hemorrhagic injection, and some old free blood is foimd in the jejunum. The Stomach is distended and contains about 500 cc. of dark bile-stained fluid. The wall is deeply injected, particularly in the cardiac portion. The wall of the duodenum is also injected. The enlargement of the lymphoid follicles in the duodenum is quite marked. The Liver weighs 3,030 gm. The surface is smooth throughout. On section the normal mark- ings are not easily distinguished and the surface appears yellowish. The gall bladder is dis- tended with about 100 cc. of rather thin bile. There are numerous enlarged lymph nodes along the common duct. Both Adrenal Glands appear on gross examination to be normal. The Right Kidney is markedly enlarged and of a soft rubbery consistency; the capsule strips with ease, exposing scattered small abscesses. On section the normal markings are only partially seen and there are numerous small (metastatic) abscesses in the pyramids and a few in the cortex. Left Kidney: The left kidney weighs 360 gm., shows just under the capsule a number of small subcapsular and cortical abscesses. Otherwise, it is similar to the right. Permission for an examination of the Central Nenous System was unfortunately not granted. The significant gross findings therefore were the enlargement and softening of the spleen and liver. The latter organs showed diffuse acute parenchymatous changes. The enlargement of the lymph follicles throughout the duodenum and swelling of the portal lymph nodes was noteworthy. A fresh fibrinous pleurisy on the left and right lung was associated with the red and gray hepatization and multiple-abscess formation in the right lung. Of particular interest Avas the bilateral metastatic descending nephritis. Microscopic examination of sections stained according to the method of Wolbach with Giemsa's stain, hematoxylin-eosin, and Sudan III revealed the following significant findings: Lungs: The portions of the left lobe lying beneath the pleura covered with partially or- ganized fibrin nets show marked filling of the alveolar capillaries with red cells. Similarly the blood vessels of the bronchioles are affected. Few alveoli are empty. Directly under the pleura the cell linings of the air sacs are desquamated and imbedded in amorphous material. Below this layer the majority of the alveoli are filled either with red cells or with masses of polymorphonuclear leucocytes and some monocytes. Occasionally an air sac contains a net- work of delicate fibrin threads and a few leucocytes, while others reveal clusters of "dust cells." In the bronchioles and bronchi either red cells or amorphous pink-staining detritus without leucocytes cover the normal mucous lining. In the deeper sections large areas of the parenchyma consist of alveoli gorged with red-cell masses or a transudate. A similar picture is presented by the right lobe, although alveoli with polymorphonuclear exudate enclosing pneumococci predominate. Located directly under the pleura or adjacent to it many alveoli form a confluent mass consisting of mononucleated cells with very few polymorphonuclear cells and some nuclear detritus. The remains of the alveolar septa are hyalinized and necrotic. In the center of these areas dense deeply stained clumps of gram-negative coccoid bacteria are present (pi. 2, fig. 3). Morphologically the organisms resemble Brucella. However, in some alveoli clumps of gram-positive diplococci and aggiegates of single cocci embedded in a mesh- work of long gram-positive and gram-negative rods are seen. The exudate surrounding the bacterial masses is relatively small and consists of polymorphonuclear cells which have phagocytized a great many of the cocci. The alveolar sacs surrounding these various types of abscess are engorged with red cells, not infrequently forming a confluent hemorrhagic pneumonic consolidation. There are no cellular reactions in the bronchioles, although pink- staining loosely arranged material is present in the lumen. The exact origin of the abscesses induced by the Brucella organism is not definitely dis- K. F. Meyer aa^ cernible, but in some sections clumps of bacilli are noted in the capillaries of the interlobular vessels. It is reasonable to suspect that they are the seeds for the mctastic-cmbolic origin of the abscesses. On the other hand, the bacterial aggregates in the alveoli of the right lower lobe represent early stages of microbial localization which would ultimately have led to small focal or even a lobular aspiration pneumonia. The extensive red and gray hepatization is unquestionably induced by pneumococci. Mediastinal and Bronchial Lymph Nodes: The general engorgement of the capillaries and the looseness of the follicles separated by dilated sinuses explains the edematous state of the organ. Cortical as well as medullary sinuses are filled with amorphous material, a few leuco- cytes, monocytes, and red cells. Pneumococci are readily demonstrable. Spleen: Engorgement of the venous sinuses with red-cell masses predominates to such an extent that wide areas of the spleen resemble large hemorrhages. Although polyblasts and lymphocytes are present, monocytes are numerous and predominate in the sinuses located luider the capsule. The Malpighian bodies withoiU germinal centers are of a loose structure; they are slightly enlarged and completely submerged in the masses of red cells. In a great many areas of the spleen nucleated cells are only noted along the trabeculae and the penicilli. Intestinal Canal: Numerous hemorrhages are present in the villi of the duodenal and jejunal mucosa. The lymphoid-tissue follicles show a loose arrangement and lack distinct germinal centers. Liver: The mosaic of the liver lobides is frequently indistinct; the sinusoids are distended and filled partly with amorphous material, red cells, occasional polyblasts, and monocytes (pi. 2, fig. 4). In the periphery of the lobule many of the hepatic cells contain fine and large fat droplets, some show nuclear necrobiotic changes while others are acidophilic and in a state of dissolution. Throughout the columns the hepatic cells show a coarsely granular or even a flaky cytoplasm; those adjacent to the central vein are in part studded with irregular coarse particles of bile pigment. The Kupffer's cells contain hemosiderin. Bacteria are not recognized despite careful search. The gall-bladder ^vall is slightly edematous; the mucosa is normal. Portal Lymph Nodes: The capillaries are markedly distended and filled with red cells. In the cortical and subcapsidar sinuses masses of red cells, polymorphonuclear leucocytes, and monocytes are present. The secondary nodules of the lymph follicles are faintly visible (pi. 2, figs. 5 and 6). Several of the medullary sinuses contain nests of epithelioid hyperplasia first described by Fabyan^ in connection with his studies on the pathogenesis of abortus in- fections in guinea pigs. An occasional cell contains one or two gram-negative coccoid rods. Kidneys: Both organs show in general the identical involvement, although the areas with abscesses are more evident in the right kidney. Aside from an engorgement of the blood vessels in the pyramids, the tubules are markedly dilated, filled with albuminous content and occasional amorphous cluinps of pigment. As a rule, the glomerular tufts are normal in size, in part filled with blood, in part collapsed, but the subcapsular space contains albuminous detritus. In several renal corpuscles of many sections examined, the capsular epithelium reveals a number of deeply stained cells protruding into the lumen of Bowman's capsule (pi. 2, figs. 7 and 8). Under high magnification these cells contain clusters or morula- like masses of gram-negative bacteria. These abnormal renal corpuscles arc in the vicinity of extensive areas of cellular infiltration which may be located directly beneath the capsule in the center of the cortex or even in the medullary substance (pi. 3, figs. 9 and 10). As a rule, the center of such a zone contains one or several transverse sections of proximal convoluted tubules with dark bluish or acidophilic cells attached to the basement membrane. These cells are also seen in the lumen of the excretory ducts. Close examination of the bluish cells re\'eals that they have enclosed in their cytoplasm irregular masses and nests of bacteria indistinguishable from typical Brucella organisms. Many of these cells have a darkly stained round nucleus while in others various stages of karyolysis attest to necrobiotic changes which have apparently taken place. The bacterial 442 Pathogenesis o£ Undulant Fever clusters neatly arranged in discs or spheres are intracellular colonies of Brucella which de- veloped in the living epithelium of the proximal convoluted tubules and the capsule (pi. 3, figs. 11 and 12). Surrounding these abnormal tubules, massive infiltrations of the stroma by monocytes, lymphocytes, plasma cells, and an occasional polyblast give the lesion a pattern which distinguishes it from the common renal abscesses induced by pyogenic cocci. Moreover, the arrangement of the epithelioid cells indicates a focal giowth incited and maintained by the activity of the bacteria (pi. 3, fig. 13). As a whole, the lesions resemble those described by Fabyan for the kidneys of guinea pigs infected with Br. abortus. The infiltration extends along and between the collecting system and compresses other tubules. On the outer zones of these areas the capillaries are greatly dilated and filled with red cells. Throughout the tubular system one encounters single or small clusters of bacteria as free masses or embedded in albuminous material and desquamated cells ^vithin the lumen. In the medullary zone an occasional collecting tubule with completely necrotic epithelial cells encloses a solid plug of gram-negative bacilli (pi. 3, fig. 14). The reactive inflammatory process surroimding these tubules consists predominantly of monocytes or histiocytes and a few polyblasts, and extends rootlike between the uriniferous structures deep into the pyramids. These cells are imbedded in a mesh of slightly thickened reticular fibers. Some of the cells harbor phagocytized fragments of nuclear material in the broad cytoplasm. The epithelium of the convoluted tubules not involved contains small fat droplets, and occasional single cells or groups of cells are acidophilic and necrotic. Bacteriological Examination of Body Fluids and Tissues Recovered at Autopsy The specimens were cultured by impressing freshly cut surfaces of the organs over several blood- or liver-agar-plates or by streaking 0.1 cc. of the fluid-triturated organs (1 gm. in 10 cc. of broth) over the media. Heart Blood: Pneumococcus Type 2 and Type IV (?); E. coli and Staphylococcus aureus. Right Lung: Metastatic abscesses: Brucella suis (about 80 per cent), pneumococci, Staphylo- coccus aureus, very few E. coli. Consolidated area: Pneumococci, hemolytic cocci and few E. coli. Mediastinal Lymph Node: Pneumococci, E. coli and staphylococci, about 4 Brucella colonies. Liver: 18-25 colonies of Brucella suis, a few E. coli (2 types not identified). Bile: 0.1 cc. of dilution 1:10 yields approximately 300 Brucella colonies or approximately 30,000 bacilli per cc. of bile and 400 E. coli. Portal Lymph Nodes: Second dilution plate: 14 colonies of Br. suis, 9 E. coli and a few colonies of staphylococcus. Spleen: Second dilution plate: 100 colonies of Br. suis, 40 hemolytic coccal colonies; third dilution plate: 20 colonies of Br. suis, few coccal colonies. One gm. of splenic tissue was emulsified in 10 cc. of broth; 0.1 cc. of the organ suspension yielded 36 colonies of Brucella suis or j,6oo organisms per gm. of splenic pulp. Left Kidney: Innumerable colonies of Br. suis. Right Kidney: Innumerable colonies of Br. suis and 4-5 of E. coli. Ileum: (hemorrhagic segment); principally E. coli but a few Br. suis. The Brucella strain isolated from the blood culture taken on November 26 and at least 10 other strains grown from organs and selected at random, grew freely aerobically, formed abundant H2S, developed readily in the presence of thionin and moderately heavily in basic fuchsin (1:50,000), and were identified with the laboratory strain Br. suis No. 80 (Meyer and ZobelP). It was specifically agglutinated by an absorbed abortus serum and it infected guinea pigs with typical granulomas which presented central aggregates of polymorpho- nuclear leucocytes. K. F. Meyer . 443 Discussion Pure Brucella abortus infections with a duration of less than three weeks have not as yet been reported. Therefore, it is indeed a rare experience to be able to study an overwhelming Brucella sepsis during its rapid clinical course as well as post mortem. Unfortunately, the Brucella disease was complicated by a lobar and lobular pneumonia in part induced by pneumococci, staphylo- cocci, and streptococci. In consequence, the lesions in the lungs were partly obscured by the supervening microbian invasion, which was in all probability of the type of an aspiration pneumonia with relatively little involvement of the bronchial tree. On the other hand, there is definite proof that the abscesses in the right lobes were at least in part of hematogenous origin, and repre- sented metastatic-embolic localization of the Brucella which circulated in very large numbers in the blood stream early in the course of the infection. Both the bacterioscopic and cultural examinations proved the predominance of the Brucella organisms in the purulent monocytic exudate. Old (Sharp^) and new (v. Albertini and Lieberherr,* and Rabson") analyses of the pathology of undulant fever mention pulmonary localizations as a complication of the disease. Hardy, Jordan, Borts and Hardy" (case 40) describe a pulmonary abscess in a patient who was infected with both suis and bovis varieties, but they were unable to determine whether it was caused by one or both of the primary infectious agents or to some secondary invasion. In all probability, the same difficulties would have arisen had the patient, D.S., survived for a slightly longer period of time. The presence of clusters of a great variety of bacteria within many of the air sacs filled with an inflammatory exudate amply attest to such a possibility. The extensive icterus, the increased weight, and the mottled appearance of the liver suggested extensive lesions in form of widespread necroses, which were readily recognized both in the periphery of the lobule and occasionally in the center of the columns of the hepatic cells. Necrobiotic processes around the central vein, as described by Lillie (see Hardy and associates,* fatal case iD and 2D, p. 77-79), Wohlwill,'' and others were relatively rare. Focalized necroses or typical "granulomas," as observed by a number of pathologists (for detailed references see Rabson^) in livers derived from human or guinea pig Brucella infections, were strikingly absent. To be sure, the extent of the liver damage hardly accounts for the icterus. This complication seems to be rare. The analysis of 125 cases (1929) and subsequently 300 cases by Hardy and his associates,"'^ also Hardy's summary in Huddleson," and the review by Sharp^ make no mention of jaundice as a symptom of ab or tus- wndnX^nt. fever. It is stated that a subicteric tinge is common and that in all severe cases there is urobilinogenuria. Ebskov and Harp0th'° in describing a case of febris undulans with jaundice interpreted the complication as a hemolytic icterus induced by the infection. Schittenhelm"^ observed icterus in one case of a series of sixty. Urobilinogen is present in the urine of one-half of the cases observed; but, since localized pain in the region of the gall bladder arouses 444 Pathogenesis o£ Undiilant Fever the suspicion of cholecystitis, it infrequently remains undecided as to whether the icterus is caused by increased bilirubin formation or insufficiency of the liver cells. Schlierbach and Wurm'^ mention a Bang infection which appeared with the symptoms of a "catarrhal jaundice." Both the direct and indirect van den Bergh reactions were positive in the blood plasma of D.S. An excess of urobilinogen in the urine would suggest that the jaundice resulted from overproduction of bile pigment. As essential factors, one might suspect the fever associated with anoxemia which depresses the function of the hepatic cells. On the other hand, the direct-reacting plasma- bilirubin was demonstrable in the blood. Thus the widespread necrosis of liver cells was followed by leakage of bilirubin from the canaliculi into the blood. Overproduction led to retention and later to hepatic damage with subsequent regurgitation. In the future the faintly marked jaundice— which may be observed in the course of a Brucella infection— should be carefully analyzed in order that the etiology and pathogenesis may be clarified. It is becoming increasingly evi- dent that the term "catarrhal jaundice" has been applied to a mixture of different etiologies. From the facts thus far available, it is clear that a moderate retention jaundice due to overproduction of bilirubin is a common symptom of brucellosis. The bile contained in the gall bladder at autopsy yielded 30,000 Brucella organisms per cubic centimeter. It is not unlikely that, just as may be the case in typhoid fever, this viscus becomes the seat of prolonged harborage and prolifer- ation of the Brucella organisms (Kennedy," Eyre," and others). In fact, in quite a few instances it has bcome necessary to remove the gall bladder to relieve the patient and to accelerate recovery (Bull and Gram;" Simpson;^* Hardy and associates;" Leavell and Amoss;" Amoss;" Mettier and Kerr;'" Menefee and Boston;"" and others). Unpublished experiments on rabbits and on guinea pigs indicate that intravenously inoculated Brucella organisms promptly appear in the bile. The larger the inoculum the greater the shedding of bacteria in the bile of the common duct or the acciunulations in the cystic bile. Thus the hemato-hepatic route is the usual one in the animal and in all probability also in man. From what is known concerning the pathogenesis of cholecystitis, it would indeed have been a miracle if this organ had escaped invasion in the presence of a blood-stream infection of over 1,700 organisms per cubic centi- meter. A careful search for embolic-metastatic foci throughout the mucosa of the gall bladder was entirely negative. However, it should be remembered that with the progress of the inflammatory reactions in the liver, the biliary ap- pendices may become involved. The persistence of the Brucella bile-carrier state in all probability depends on the degree of the inflammation of the gall- bladder wall. It was noted that the portal lymph nodes were definitely enlarged and hyperemic. Significant lesions in the form of a desquamative process in the cortical sinuses and the early aggregations of mononuclear cells and a few K. E Meyer 445 polyblasts in the medullary sinuses appear to be attributable to the action of the Brucella organism. Through culture and through bacterioscopic ex- amination the heavy invasion of the nodes was definitely proved. Just as in typhoid fever, so in brucellosis a definite relationshi}) exists between these cellular elements and the bacteria. Perhaps the phagocytic activity of these cells is important in the immunity mechanisms. Von Albertini and Lieber- herr* have noted the frequent involvement of the portal and retroperitoneal lymph nodes in Bang's disease, and although the duration of the disease was approximately five months in one of their cases the lesions were quite similar to those seen in the organs of D.S. Definite epithelioid nodules, which replace the lymphoid structures, are missing, but the aggregates of monocytes inter- spersed by lymphocytes and a few polyblasts are unquestionably precursors of the focal or diffuse structures found in the lymph nodes, liver, spleen, bone marrow, etc., of human brucellosis. These nodular reactions, with which every pathologist is familiar, since they were described by Smith and Fabyan,^ by Fabyan,-^ and by Smith"^ were first recorded by Loffler and v. Albertini"* and shortly thereafter by WohlwilF in a human autopsy specimen from proved Bang infections. Subsequently, their presence has been amply confirmed by many investigators (see Rabson^). However, not every fatal Bang infection exhibits this nodular reaction. In fact, as autopsy reports and microscopic tissue examinations accumulate, it becomes apparent that these focal in- flammatory or proliferative processes, although characteristic, and in their later stages diagnostically significant for Brucella infections, are the excep- tion rather than the rule. That they are the resultants of reactions which take place in the reticulo-endothelial system in response to the phagocytosis of bacilli in the blood or lymph streams, there is little doubt, but the im- munobiological factors responsible for their formation are entirely unknown. Rossle"' "* interprets these cellular aggregates as an expression of allergic re- actions. Until the mechanism of the immunity in Brucella infection is really understood, it is advisable to abstain from premature attempts to explain these nodular lesions. While their genesis remains a fascinating study subject to experimental analysis, it is apparent that these structures contribute an important share to the well-known latency and chronicity of undulant fever. In this connection, it should be recalled that dead brucellas or their protein may produce similar lesions in the sensitized hosts; consequently one may encounter nodules which are bacteriologically sterile. It is reported with increasing frequency that the hepato-licnal syndrome governs a Bang infection. Aside from the pathology in the liver, the most obvious and general gross changes, particularly in cases of long standing, is the enlargement of the spleen. On account of the rapid course of the infec- tion in the patient, D.S., the splenic organ showed merely the early stages of an inflammatory reaction, accompanied by hemorrhagic engorgement which sub- merged the follicles. The diffuse reticulo-endothelial hyperplasia might pos- sibly be interpreted as early stages of the "tuberculoid reaction" so frequently 446 Pathogenesis o£ Undiilant Fever observed in the spleen in the latter stages of undulant fever. As a sequel of the inflammatory process, a Bantilike disease, accompanied by atrophic liver, cirrhosis with ascites, and sub-icterus, has been noted (Habs," Biirger,"^ ™ Loffler,*" Schottmiiller"). The gross lesions in the intestinal tube suggested, aside from the few extravasations in the villi, some involvement of the lymphatic apparatus. Despite the examination of many sections, definite hyperplasia of the reticulo- endothelium of the follicles was not discovered. Just as in the spleen, the germinal centers were lacking. Of all the changes, those noted in the kidneys are the most interesting. Localized infiltrations involving the interstitial tissue between the tubules have been described by v. Albertini and Lieberherr,* Cortese,^' de Giorgi,''^ and Gregersen and Lund.^* Equally, necroses of the tubular epithelium lead- ing to nephroses are on record (Bruce;'^ De la Chapelle;^'' Hardy and asso- ciates,* case iD). Clinical data furthermore attest to the occasional involve- ment of the kidneys. Albuminuria with hyaline and occasionally granular casts, even leucocytes and red blood cells, indicate either toxic effects or localization of the specific infection in the urinary tract. The Brucella organ- ism may be isolated from the urine up to 12 months after the onset of a melitensis infection, or the continuous discharge of Br. suis in the bladder of a specifically allergic individual may give rise to a hemorrhagic cystitis without significant cystoscopic and pyelographic findings (personal observa- tions). As a whole, the participation of the kidneys in the course of melitensis undulant fever infection may involve at least 10 per cent of the patients (Dun- can and Whitby^^); on the other hand, a mere dozen of a series of 200 patients with melitensis brucellosis seen by Cantaloube^^ yielded albumin in any signifi- cant amounts, while Simpson^'^ states that only traces of albumin are found in abortus infections. It will be shown below that the absence of chemical or physical damage does not preclude a bacilluria. The students of the experimental disease in guinea pigs frequently found the kidneys of the animals injected with Br. bovis diseased. According to Fabyan,^ 29 per cent of these animals showed whitish to gray nodules in the cortex. Radiation toward the medulla was not observed, although the color- less appearance of the renal structures indicated a diffuse process. Microscopi- cally, the focal areas were composed of epithelioid and lymphoid elements, compressing and destroying the convoluted tubules near the glomeruli. In other instances, these "granulomas" were in the vicinity of blood vessels. As far as published records indicate, no attempt has been made to explain the genesis of these lesions, although the examinations of chick embryos infected with Br. suis reveals granulomatous foci consisting of proliferations of endo- thelial cells remarkably similar to those noted in human kidneys. The find- ings made on the sections of the kidneys removed at autopsy from the body of D.S. may shed some light on the evolution of the renal pathology in brucellosis. K. F. Meyer 447 First and foremost is the intracytoplasmic multiplication of the bacteria in the epithelium of Bowman's capsule and the convoluted tubules. It was thought that the propagation took place in necrotic cells, but, on closer scrutiny of many sections, invasions and viability were equally present in cells Avith intact vesicular or dense nuclei. In fact, the peculiar intracellular giowth differs in no way from that first described by Theobald Smith^" in the chorionic epithelium of the calf, or of the goat (Meyer*"), and more recently by Goodpasture and Anderson" and b) Buddingh and Womack'" in the vascular endothelium and cells of mesodermal origin of the growing chick. These authors justly concluded that the ectodermal epithelium and mesodermal cells serve as suitable host cells and media for invasion and propagation for Br. abortus and Br. suis. Experiments of a similar nature conducted with the Br. suis strain isolated from D.S., and two years before the paper of Buddingh and Womack had appeared, amply confirmed the observations of Goodpasture and Anderson, but revealed the important fact that the endothelial cells lin- ing the columns of hepatic cells and the glomeruli of the nephros in the embryos sacrificed on the fourth day were frequently crowded with specific bacteria as intracytoplasmic inclusions. The microscopic pictures were in- distinguishable from those seen in the human renal cells. This selective intra- cellular parasitism in mesenchyme cells of various organs is doubtless of greatest significance in the pathogenesis of Brucella infections. Up to the present time, the available observations on undulant fever have indicated neither the portal of entry of the Brucella organisms into the tissues of the animal or human host, nor the sites of multiplication of these organisms after in\ asion has taken place. There is no proof that brucellas multiply in the blood stream, although they will grow freely in necrotic tissue (chick- chorion allantoic membrane; abscesses in lymph nodes and epididymis of guinea pigs). Their presence in macrophages and polymorphonuclear leuco- cytes is justly attributed to phagocytosis, there being some indications that under certain conditions even the cytoplasms of these cells may serve as a medium for propagation. Little attention has been paid to the possibility that the epithelial cells of the proximal convoluted tubules of Bowman's capsule, and of the chorionic epithelium may possess phagocytic power. For the locali- zation in the latter, Theobald Smith believes that the bacilli pass from the blood vessels into the uterine cavity where they are rubbed into the substance of the cell by pressure. Once in the cytoplasma, they find conditions favorable for multiplication and protection against phagocytosis. As far as the renal localization is concerned, two possibilities may be envisioned: {a) the bacilli, which pass the glomerulus without causing visible damage, may be pressed by the urinary filtrate into the epithelial cells, or {b) since a peculiar group of renal cells-the lining of Bowman's capsule and the proximal end of the con- voluted tubes-are involved, the possibility of an end result of an excretory function should not be overlooked. The visible phenomenon of intracyto- plasmic parasitism in the epithelium may be the residt of "resorption" or 448 Pathogenesis o£ Undulant Fever "secretion." Knowledge concerning the physiology of these groups of cells is largely speculative. The localization, however, is probably not a mere co- incidence, since it is well-known that Sporozoa of the character of coccidia not infrequently are present in the same cellular elements as the Brucella organism. Apparently, the nutritional conditions in the epithelia are suitable for a diversity of parasites, which somehow enter the cytoplasm through motility of their own or through other physical forces as yet poorly understood. As a sequel to intracytoplasmic proliferation, the majority of epithelial cells undergo necrosis. The bacterial masses are discharged into the collective tubules where they form casts, and in turn cause death of the tubular lining. A reactive inflammation extends along the uriniferous tubules deep into the medulla and even into the calyces. The cells forming the infiltrative reaction are mostly mononuclear histiocytes and lymphocytes; some with broad cyto- plasmic margins have ingested nuclear detritus and some bacilli. This type of medullary localization is merely a part of the metastic elimination yiephritis. The possibility exists that renal abscesses or inflammatory infiltrations may be the sequel of hematogenic embolic processes, which are so commonly seen in coccic infections, and described for lesions observed in guinea pigs in- fected with Br. abortus or Br. suis. A careful scrutiny of many sections from ten separate portions of the kidneys of D.S. failed to demonstrate connections with emboli. The entire process is confined to the uriniferous tubules, and ultimately to the intracytoplasmic growth and necrosis of the epithelial cells. What the outcome of these lesions may be is well illustrated by v. Albertini and Lieberherr,* who foimd in the fifth month of a Bang's infection massive interstitial granulation tissue composed of large phagocytic granuloma cells. Unfortunately, no information relative to the bacteriology of the urine of this particular case has been published; it is noted that the diazo reaction was strongly positive. It would have been of interest to know whether the focal lesions in the kidneys were closed or whether they discharged Brucella occasionally or intermittently. According to Huddleson," Brucella abortus has been cultured very few times from the urine of infected human beings. On the other hand, Horrocks^"* succeeded in culturing Br. melitensis 39 times from the urine of 13 different patients. He reports that the organism has not been isolated from the urine earlier than the fifteenth day or later than the eighty-second day of disease. The actual number passed varied between 4 to 596 organisms per cubic centimeter. Occasionally, a trace of albumin was noticed, but no physical or chemical changes common to all the urines and in- dicative of the passage of Br. melitensis has been observed. As a rule, the dis- charge of the bacilli was intermittent and occasional, while in very few cases it was continuous over a period of two weeks. The observations were made on cases of the undulant, intermittent, and irregidar types of melitensis fever. Renewed interest in the bacteriological examination of the urine in brucellosis infections due to Br. abortus and Br. suis might indicate that the discharge of the specific bacteria is not so rare as published statements would imply. K. F. Meyer 44g Fortuitous circumstances permitted an exhaustive bacteriological exami- nation of the organs of D.S. at the time of autopsy. The quantitative estima- tions reveal more definitely than any other evidence the septicemic-metastatic character of the infection. Every organ, in fact every tissue fragment, yielded Br. suis, sometimes in such numbers that accurate counting was impossible. The absence oi Brucella in the heart blood must be ascribed to the overgrowth by the pneumococcus in the rich culture afforded by the clot in the right ventricle. Attention must be directed to the estimation of 3,600 organisms per gram of spleen or approximately 1,512,000 bacteria for the entire organ which, in all probability, is due to the presence here of the main reticulo- endothelial system and its "filteration" of the Brucella from the blood stream. Such a limited storehouse of bacteria could not have served as the seedbed for the blood stream. In all probability the distributions of the organisms is uneven and the number of bacteria calculated for the entire organ is probably too low. However, it should be remembered that in typhoid fever the con- tinuous bacteremia is, as a rule, maintained by the focal lesions in the bone marrow. The same is suspected for a Brucella infection. Unfortunately, the possibility had not been considered and therefore no cultures were prepared from the marrow of D.S. It certainly should not be missed in future autopsies. The literature on undulant fever due to Br. abortus or suis contain 44 re- ports on supposedly fatal infections attributable to the specific organisms. In i^ of the reports no microscopic-anatomical data are presented. The histories of 15 cases together wath 8 observations of their o^vn have been analyzed by v. Alber- tini and Lieberherr.^ In addition, data on 7 cases of Bang's infection (one by De la Chapelle^° was caused by Br. 7nelite7isis) are detailed in the American literature (Rabson°). A condensed analysis definitely indicates that Brucella infections due to the bovis or suis types are very rarely fatal. Of the total num- ber of 44 only 7 were uncomplicated deaths directly attributable to the specific organisms. In the remaining 37, death was due to some complication such as embolism, uremia, tuberculosis, hemorrhages, rupture of the spleen, hepatic cirrhosis, pneumonia, recurrent endocarditis, etc. Invariably in typical Bru- cella suis or abortus deaths, a definite sepsis was demonstrable. To attribute a death to a Brucella infection because the patient's serum gave immunological reactions suggestive of undulant fever, without an autopsy followed by micro- scopic study of the tissues and detailed bacteriological examination of the organs, is not justifiable in the light of available experience. No effort should be spared to secure autopsies on supposedly fatal cases of brucellosis. Knowledge relative to the pathogenesis of the Brucella infections in the aberrant host— man— is entirely inadequate. The pathways of infection are supposedly proved. One suspects the intestinal and the percutaneous routes. The clinical course manifests itself in the form of an intermittent or undulant fever. By analogy with malaria, spirochetosis, and endocarditis, the bouts of fever presuppose repeated invasions of the bacteria into the blood stream. These seedings come, according to Wohlwill." from an\ one of the reticulo- 450 Pathogenesis of Undulant Fever endothelial beds in the organs. Loffler and v. Albertini'' specifically place the focal processes in the spleen and liver. This interpretation of the pathogenesis is largely based on the fact that the characteristic "tuberculoid, nodular" lesions or "granulomas" are principally noted in these organs and the culti- vation of the organism from these tissues is frequently successful. Their pres- ence in the lymph nodes is believed to be secondary and not the resultant of hematogenic invasion. The explanation by Loffler-' has a great deal of plausi- bility. In uncomplicated Brucella deaths the spleen is regularly enlarged; furthermore, the majority of clinicians have found the spleen palpable in varying degrees. The German data mention this sign in 80 per cent of the cases; on the other hand, Simpson,^' and Hardy and associates* found a splenic tumor in only 30 per cent of their observations. Thus, there is no justification for considering the splenic tumor as an obligatory sign of brucellosis. The liver is enlarged even less frequently (German data 20 per cent; 4 per cent in Simpson's cases). No correlations between the enlargement of these organs and the blood-culture findings are available. Nothing is known concerning focal seedbeds in the bone marrow; both Wohlwiir and Wegener" have found epithelioid nodules and even necroses. Even the best culture methods used repeatedly and by experienced workers have yielded negative blood cultures. It is, therefore, not unlikely that other factors than the bacteremia are responsible for the febrile reactions. Since true toxins of the Brucella organism have not as yet been demonstrated, one has to think of the proteolytic degradation of the harmless albuminoid body substances to poisons or the formation of febrigenic substances to poisons from the tissties of the infected host through autolysis. These poisons in turn gen- erate new foci of inflammation, and thus fever-inducing factors are created without the actual proliferation and dissemination of the specific bacteria which initiated the disease. A series of febrile elevations of the body tempera- ture may not be accompanied by or induced by a bacteremia. Finally, there is the possibility that the symptoms are manifestations of the allergic state which always develops in the course of a brucellosis. The paren- teral presence of the bacterial elements as antigens stimulates the production of antibodies in the tissues, and these in turn come in contact with the bacterial protein which is liberated during the disintegration of the bacteria in the various phagocytes. Through interaction of these two components a disease- producing agent may be formed. Obviously, it is subject to discussion as to how these substances are formed: the combination may produce the poison, the disintegration of the bacterial protein may liberate an injurious substance, or the evolution of the antibody per se may represent the noxis. In Brucella infection, just as in tuberculosis or in fungus disease, the allergy factor plays an important role in the mechanism of immtuiity, and a fair portion of the symptomatology may be conditioned by this altered state of reactivity. All these attempts to explain the pathogenesis of a Brucella infection leave out of consideration the primary question: Does the organism lead an intra- K. F. Meyer 4^ 1 cytoplasmic existence during its parasitism and, if so, in what cells? Compara- tive studies of infection of the goat udder, the chick embryo, and now of the tissues of man strongly suggest that the mesenchymal and possibly the ecto- dermal layers are host cells for the brucellas. Very acute infections lend them- selves to a demonstration of this stage of the parasitism. With the progress of the disease, the phagocytic activities in the reticulo-endoihclial beds, the in- flammatory reaction to the necroses in the organs damaged by the bacteria, or the split products of the allergic inflammation create microscopic pictures in which the early phases are either not discernible or so widely scattered that it is impossible to find them without examining serial sections. In the interest of a clearer understanding of the pathogenesis, it is obviously desirable to study acute Brucella infections exhaustively. In the subacute or protracted disease stages one finds either no lesions (certain questionable au- topsy reports indicate absence of organ changes while a few show specific evidence of invasion) or the "tuberculoid nodules" or "granulomas," which may be interpreted as characteristic. The factors which guide the formation of these lesions are unkno^vn. They have all the markings of a defensive cellu- lar mechanism and raise the following questions: Is it possible that the poly- morphonuclear leucocytes ingest the free bacteria under the influence of the opsonins and then become impotent, their place being taken by the monocytes, which rapidly aiTange themselves into nodules in order to destroy the bacteria? Do the brucellas, in part changed by the antibodies, lose their invasiveness and merely act to incite specific granulomas as a foreign-body reaction? Are these structures in some way connected with allergy? And finally in what man- ner are the capsules of the brucellas related to these processes? (See Mickle.*°) The microscopic findings in the kidneys and the lymph nodes of D.S. have shown that in the vicinity of the bacterial proliferation, either in the living tubular epithelium or the necrotic cellular detritus, an intense proliferation of the reticulo-endothelial cells took place. In the kidney this has occurred in the intertubular connective tissue. The cellular types, which possess all the properties of phagocytes, resembled the elements one finds in typhomas; they are monoblasts— mono- or histiocytes— intermingled with lymphocytes. The reasonableness is great that these lesions represent the early stages of the granulomas, which through necrobiotic processes finally reach the structural appearance seen and described in the majority of fatal Bang infections sub- jected to autopsy within the first to the sixth month. AVhen the infiltration is very large and the necrobiotic colliquation is extensive, typical purulent me- tastases in a diversity of organs may be encountered. Undulant fever due to Br. abortus or suis is, in ilic majority of instances, a relapsing bacteremia which may be interwoven with allergic manifestations. Although the principal place of proliferation may be the spleen, it is apparent that the bone marroAV deserves consideration. In a small percentage of infec- tions, metastatic foci are formed in the spleen, liver, the lymph nodes, kidneys, sex organs, bones, and central nervous system. Xecrobioiic processes in these 452 Pathogenesis of Undulant Fever metastases may lead to suppurative processes. Fatal septicemic infections are exceptionally rare. Finally, it is not out of place to reflect briefly on the significance of the intra- cytoplasmic parasitism of Brucella. If this phenomenon can be proved to be as common as is indicated in the growing chick and in the few observations here recorded, the concepts of the immunity mechanism, which places the entire weight on the antibodies, may require a fundamental revision. In fact, the well-known realization that it is impossible to create any noteworthy degree of acquired resistance with killed vaccines might possibly receive proper ex- planation. Just as in tuberculosis, the degree of protection afforded is pro- portional to the extent of the primary lesion or lesions. Only a true infection can produce extensive lesions in which a great many cells of the reticulo- endothelial system participate. Furthermore, these cellular elements may have had intimate contact with the Brucella organism. But what is more important is the antigenic structure. There is every reason to suspect that the antigens, which develop as a result of the intracellular parasitism, may in some respect be different from those generated in artificial media. A true infection will, in contrast with the injection of culture antigen, produce immunity. More- over, it is equally probable that the susceptible mesenchyme cells and their descendants may be rendered immune only through infection or temporary intracytoplasmic parasitism and not through mere exposure to soluble an- tigens. There is no doubt that the viability and in particular the latency— one of the most important manifestations of a Brucella infection in cattle, goats, and hogs— is intimately connected with the intracellular development cycles. Viewed from these angles the Brucella problem, so hopelessly bewildering, deserves renewed critical inquiries and experiments. Summary (i) The pathological and bacteriological findings made on an acute malig- nant infection due to Br. suis No. 80 in a laboratory-animal caretaker is re- ported. The circumstances surrounding the infection strongly indicate the ingestion of a very heavy suspension of the specific bacteria. According to the positive allergic skin reaction and the strong opsono-cytophagic response (index 6:88) the patient had been considered immune to brucellosis. (2) The metastatic lesions in the kidneys were not embolic but due to the elimination of brucellas through the glomerulus. The bacteria developed in the epithelium of Bowman's capsule and the proximal end of the convoluted tubules. The significance of the intracytoplasmic parasitism is discussed. (3) The genesis of Brucella infections is critically analyzed and suggestions for further studies are presented. REFERENCES i.Fabyan, M.: Jl. Med. Resrch. 28:81, 1913. 8.Meyer,K.F., and Zobell, C. E.: Jl. Infect. Diseases 5 1:72, 1932. 3. Sharp, W. B.: Archs. Pathol. 18:72, 1934. K. F. Meyer 4^^ 4. V. Albeitini, A., and Lieberherr, W.: Frankfurt. Ztschr. f. Pathol. 51 :6g, 1937. 5. Rabson, S. M.: Amer. Jl. Clin. Pathol. 9:604, 1939. 6. Hardy, A. B.; Jordan, C. F.; Borts, I. H., and Hardy, G. C: Natn. Inste. Hlth. Bull. no. 158, 1930- 7. Wohhvill, F.: Virchows Arch. f. pathol. Anat. u. Physiol. 286:141, 1932. 8. Hardy, A. V.: Jl. Amer. Med. Assn. 92:853, 1929. 9. Huddleson, I. F.: Brucellosis in Man and Animals (New York: 1939) . 10. Ebskov, C, and Harp0th, H.: Ugcskr. f. Laeg. 37:872, 1930. 11. Schittenhelm, A.: in Hanclbuch der Inncren Medizin, ed. by L. Mohr and R. Staehelin, I (3d ed.; Berlin: 1934), 985. 12. Schlierbach, P., and Wurm, K.: Deutsch. med. Wchnschr. 62:888, 1936. 13. Kennedy, J. C: Jl. Roy. Army Corps 6:623, 1906. 14. Eyre, J. W. H.: in Handbuch der Pathogenen Microorganismen , ed. by W. Kolle and .\. von Wassermann, IV (2d ed.; Jena: 1912), 421. 15. Bull, P., and Gram, H. M.: Norsk mag. f. laegvidnskjin. 9: 1026, 1911. 16. Simpson, W. M.: Amer. Jl. Surg. 7:597, 1929. 17. Leavell, H. R., and Amoss, H. L.: Amer. Jl. Med. Scis. 181:96, 1931. 18. Amoss, H. L.: Internatn. Clins. 4:93, 1931. 19. Mettier, S. R., and Kerr, W. J.: Archs. Internal Med. 54:702, 1934. 20. Menefee, E. E., and Poston, M. A.: South. Med. Jl. 31:1061, 1938. 21. Smith, T.. and Fabyan, M.: Cntrlblt. f. Bakteriol., Parasitnk. u. Infektnskrankhn., I. Abtl. 61:549, 1912. 22. Fabyan, M.: Jl. Med. Resrch. 26:441, 1912. 23. Smith, T.: Jl. Exper. Med. 49:671, 1929. 24. Loffler, W., and v. Albertini, A.: Krankheitsforschung 8:1, 1930. 25. Rossle, R.: Miinchen. med. Wchnschr. 80:5, 1933. 26. Rossle, R.: Wien. klin. Wchnschr. 45:609, 648, 1932. 27. Habs, H.: Klin. Wchnschr. 7:453, 1928. 28. Biirger, M.: Miinchen. med. Wchnschr. 75:754, 1928. 29. Burger, M.: Zntrlblt. f. inner. Med. 49:784, 1928. 30. Loffler, W.: Wiirzburg. Abhdln. a. d. Gsmtgbt. d. Med. no. 11, 1930. 31. Schottmiiller, H.: Klin. Wchnschr. 1 1:905, 1932. 32. Cortese, F.: Clin. med. ital. 66:774, 1935. 33. de Giorgi, L.: Gior. di clin. med. 18:565, 1937. 34. Gregersen, F., and Lund, T. M.: Hospitalstidende 74:349, 1931. 35. Bruce, D.: Ann. Inst. Pasteui (Paris) 7:289, 1893. 36. De la Chapelle, C. E.: Amer. Heart Jl. 4:732, 1929. 37. Duncan, J. T., and Whitby, L. E. H.: in A System of Bacteriology in Relation to Medicine, issued by the Medical Research Council [Of Great Britain], V (London: 1930; 388-426), 398. 38. Cantaloube, P.: La Fievre de Malte en France (Paris: 1911). 39. Smith, T.: Jl. Exper. Med. 29:451, 1919. 40. Meyer, K. F.: Jl. Amer. Veterin. Med. Assn. 39:286, 1935; and supplementary unpublished observations. 41. Goodpasture, E. W., and Anderson, K.: Amer. Jl. Pathol. 13:164, 1937. 42. Buddingh, G. J., and Womack, F. C: Jl. Exper. Med. 74:213, 1941. 43. Horrocks, W. H.: in Reports of the Commission appointed by the Admiralty, the War Office and tlie Civil Government of Media for the Investigation of Mediterranean Fever, I (London: 1905), 21. 44. Wegener, T.: Cntrlblt. f. allgm. Pathol, u. pathol. Anat. 64:33, 1935-36. 45. Mickle, W. A.: Jl. Infect. Diseases 66:271, 1940. EXPLANATION OF PLATES PLATE 1 Photographic records ol blood plates seeded with (fig. i) o.oi cc. of bile and (fig. 2) 0.01 cc. of 1 gin. of triturated right kidney in 10 cc. of broth. Specimen obtained at autopsy of D.S. C454I] Plate 1 [455 II PLATE 2 Fig. 3. Section of right lung lobe. Small abscess due to Brucella surrounded by hemorrhagic pneumonic consolidation. Zenker fixation and Giemsa stain, x 74. Fig. 4. Section of liver with distended sinusoids filled with amorphous mate- rial and necrobiotic changes in the hepatic cells. Formalin fixation and hema- toxylin-eosin stain. X 74. Fig. 5. Section of portal lymph node Avith medullary sinus filled with mono- cytes. Zenker fixation and Giemsa stain, x 74. Fig. 6. The same section; medullary sinus. X 388. Fig. 7. Section of right kidney. Glomerulus with deeply stained capsular epi- thelium protruding into the lumen of Bowman's capsule. Zenker fixation and Giemsa stain. X 98. Fig. 8. Section of the capsular epithelium; intracytoplasmic growth of brucel- las. Zenker fixation and Giemsa stain, x 388. 1:456: f ff.,U k^ i^f. '8 Plate 2 L 457 II PLATE 3 Fig. 9. Section of right kidney. Proximal convoluted tubules with deeply stained cells; the abnormal tubules are surrounded by massive infiltration of stroma by monocytes, plasma cells, lymphocytes, and polyblasts. Zenker fixation and Giemsa stain, x 98. Figs. 10, 14. Epithelial cells of the tubules filled with intracytoplasmic colonies of brucellas. Some of the cells are still normal, others necrobiotic. x 777. Fig. 11. Section through renal tubule with epithelia serving as host cells for brucellas, x 388. Fig. 12. Necrotic collective tubule plugged with masses of Brucella. Zenker fixation-hematoxylin stain, x 388. Fig. 13. Focal arrangement of monocytes surrounding tubule with cells con- taining brucellas. Zenker fixation and Giemsa stain, x 510. C458] i^^ '» , *»»C,, ^' ,10 4 . > n.' i»* • » # • ff '^Z 'i , 1 %14 Plate 3 [1459] FRENCH MEDICAL EDUCATION AS A LEGACY FROM THE REVOLUTION By J. M. D. OLMSTED From the DIVISION OF PHYSIOLOGY, UNIVERSITY OF CALIFORNIA BERKELEY, CALIFORNIA PLAN GENERAL D E L'ENSEIGNEMENT DANS L'^COLE DE SANTlfe D E PARIS. Imprtme par ordrc du Comhe d^Imtruct'iotl fublique de la Convention NationaU» De rimprimerie de Ballard fils,rue Jacque*, n*. 14a An in de la Republicjuc, Photograph of the title page of the pamphlet of 1794. FRENCH MEDICAL EDUCATION AS A LEGACY FROM THE REVOLUTION THE LEADERS of the French Revolution were nothing if not thorough in their attempts to root out all evidences of the ancien regime. Before the Revolu- tion medical education in France had been in the hands of faculties and colleges of medicine in various cities throughout the country from Rennes in Brittany to Orange in Provence. The medical school having the best or- ganization for teaching was not, as one might have imagined, located in the capital, but in Montpellier. Nevertheless, Paris was undoubtedly the great center for medical practice in France, and the provincial cities followed its lead. The Paris Faculty of Medicine included all the physicians of the city, and each year they elected certain of their number to give the public lectures. It was perhaps because the lecturer felt this to be only a temporary task, soon over, that official medical instruction had come to be perfunctory, often con- sisting in nothing more than a reading from out-of-date textbooks like that of Haller in physiology, of Boerhaave in medicine or of Hevin in pathology. Students, therefore, generally preferred to attach themselves to some private teacher who could give a more interesting exposition of the subject, or to one who had made a name for himself in the practice of medicine. Surgery was an entirely separate profession from that of medicine, and in Paris was officially taught by the College of Surgery independently of the Faculty of Medicine. In spite of the fact that the courses in anatomy, physiology, therapeutics, and even operative procedure might last a scant three months for students of sur- gery, it is claimed that these students were better taught from the point of view of practical instruction than students attending the more protracted series of lectures on the same subjects at the Faculty of Medicine. In both cases official instruction in the art of healing still proceeded along the lines laid down in the sixteenth century and had become lax and ineffective. The law- makers of the Revolution changed all this with a stroke of the pen. The Convention, in decrees of March 8 and September 15, 1793, abolished all schools, colleges, and universities, ordered all their possessions sold and their endowments confiscated. This drastic purge included the Faculties of Medicine and Colleges of Surgery, and seals were put on the closed doors of the buildings formerly occupied by them. The idea was to make education free. This was in itself a laudable enough intent, but the means adopted to bring about the desired result led merely to license and confusion. The historian Guizot, commenting some twenty years later on the effects of these measures, said, "Ignorance and disorder gained the upper hand." Of the whole French system of education the primary school alone survived, and this in spite of Jean Jacques Rousseau's advice to give a child no instruction whatsoever until he reached the age of twelve years, when the clean, unsullied page of the young 1:463] 464 A French Medical Legacy mind would all the more readily take the imprint of directions for conduct worthy of republican citizenship. It was the needs of the army which first brought home the necessity for sys- tematic instruction beyond the primary-school age, and the first institutions of higher learning to be thought of by the Revolutionaries were schools for training engineers and officers. The Central School for Public Works was de- creed March 11 and the School of Mars on June 6, 1794. Teachers were also needed, and on October 30 came a decree establishing a Normal School, a center for instruction "in the art of teaching morals and forming the minds of young republicans in the practice of public and private virtues." Finally, on December 4 (14 frimaire an 3, according to the Revolutionary calendar) the Convention made provision for three Schools of Health, one in Paris, one in Montpellier, and one in Strasbourg. The School of Mars and the Normal School were very short-lived, their existence being only a matter of a few months, but the School of Public Works became the well-known Paris Ecole Polytechnique, and the three Schools of Health became the leading medical schools in France, although one of them has been forced to change its nation- ality three times. To Antoine Francois Fourcroy belongs the honor of setting in motion the legislation which gave rise to present-day French medical education. He him- self had obtained his medical degree in 1780 with great difficulty because of poverty. The 6,000 livres necessary for the diploma had been contributed by friends of the celebrated anatomist, Vicq d'Azyr, who boarded with young Fourcroy's family. Fourcroy had welcomed the Revolutionary movement and the reforms it promised, but he was averse to entering the wild arena of prac- tical politics and at first refused to accept any office. In spite of his protests he was elected a member of the Convention— one of the few physicians in this body— and here he devoted himself almost entirely to questions concerning education. Convinced of the necessity of a supply of physicians, chiefly for the army, he consulted Prieur, the member of the Committee on Public Safety in charge of the teaching of sciences and arts, asking him to recommend some- one capable of collaborating in mapping out a scheme of organization for medical education. The name of Francois Chaussier was suggested. Chaussier was not a Parisian, but had been prominent in medical circles in the provincial city of Dijon, holding appointments there not only as Surgeon of the Prisons and Physician of the Hospitals but also as Professor of Chem- istry, and giving courses in anatomy and legal medicine. Because he sym- pathized with the idea of the Jacobins regarding the centralization of all power in Paris he proposed the establishment in that city of a single Central School of Health. Jacobin principles, however, had just passed into disfavor, and members of the Convention agreed that similar schools should be established at Montpellier and Strasbourg. Fourcroy concurred and the decree was so worded. It is curious to note that when the government came to publish the text of the decree it prefaced it by a copy of Chaussier's report advocating a J. M. D. Olmsted ^65 single Central School of Health in Paris. A few footnotes were added to the effect that the recommendations contained in the report were applicable to three schools as well as to a single one. The telling argument m this report was, of course, the need for "health officers in military hospitals and camps," for it was stated that "six hundred such officers have perished in the last eighteen months. If it is a glorv for them, since they died serving their country, it is a necessity for the republic to replace this loss." Chaussier went on to point out that opportunity was at the same time afforded to organize a complete system of the art of healing in a way never before attempted. To remedy one of the most glaring defects of pre- Revolutionary medical instruction, namely, the mere reading of obsolete texts, he proposed a startling innovation— lectures were to be supplemented by "exercises on the part of the students, chemical experiments, anatomical dis- sections, surgical operations, use of apparatus. Little to read, much to see, much to do, such shall be the basis of the new teaching . . . Observation at the bedside of patients shall become one of the chief principles of this teach- ing." A second recommendation of equal importance for medical education was to be found in his suggestion that "medicine and surgery are two branches of the same science," and should therefore be taught together to all prospective health officers.* The decree as passed by the Convention embodied Chaussier's recommenda- tions, which had been transmitted through Fourcroy. Article i stated the pur- pose of the measure: "There shall be established a School of Health at Paris, at Montpellier, and at Strasbourg; these three schools shall be destined to produce officers of health for the service of the hospitals." Article 3 outlined the general subject matter to be taught, such as "the signs and symptoms of disease," etc., and Article 4 contained the important rider, "Besides this first part of the teaching, students shall practise anatomical, surgical, and chemical operations; they shall observe the nature of diseases at the bedside of patients, and shall follow their treatment in the hospitals near the schools." Article 5 provided eight professorships at Montpellier, six at Strasbourg, and twelve at Paris, together with an equal number of associates. Permanent professorships would ensure continuity in the mode of teaching, lack of which had been one of the weaknesses of pre-Revolutionary days. Article 13 showed that Fourcroy * The idea tliat surgery should be coequal uith medicine did not originate with C:haussier, but had gradually been taking shape for several years. In 1790, that is, shortly after the down- fall of the monarchy and the establishment of the National Assembly, but before the dissolu- tion of the Faculties of Medicine and Colleges of Surgery, all learned societies were required to make a report to the Assembly on changes in their constitutions necessitated by the new order of society. The Royal Society of Medicine was one of the group and in their report to the Assembly they devoted several pages to a plea that surgeons be required to start their professional training with as good a preliminary education, and that their professional studies be as extensive, as those demanded of physicians. Surgery would thus be placed on a par with medicine. This point of view was clearly stated in the following sentence: "We beg those who are still astonished at this conclusion to reflect that since the division of disease into internal and external is vicious, the separation of medicine and surgery, which depends on such a divi- sion, can no longer be maintained." 466 A French Medical Legacy remembered his own difficulties in obtaining a medical education, for there was provided a yearly allowance of 1,200 livres to each of the three hundred chosen students for a period of three years, this being the length of time allotted for a complete course of instruction. Two months later the Committee on Public Instruction, headed by Four- croy, authorized the printing of a second small pamphlet entitled "General Plan of Teaching in the Paris School of Health." This pamphlet proved to be virtually an announcement of courses, with a schedule of class hours. It con- tains, however, a much fuller description of the contents of the courses and their aims than is usually to be found in the modern university catalogue. Instruction was classified under two headings, "Permanent courses" and "Non-permanent courses or semesters." The former were to include the clin- ical instruction in the hospitals which each student must attend daily from the very beginning of his student days to their end; the latter were to include twelve courses in special subjects taught by the twelve newly appointed pro- fessors and their twelve associates. An academic year lasted a full calendar year. The students' hours were arranged as follows: up to 10 a.m. (hospital rounds began at 7 o'clock) and during his spare hours, he must spend his time in one or other of the hospitals the year round. Three hospitals had been selected for teaching, and students were to be assigned to each one for a period of four months, thus going the entire round in a year's time. In the winter semester, beginning September 24, first-year students at 1 o'clock daily, "except on the tenth day consecrated to rest," listened to lectures in anatomy-physi- ology. These two subjects were taught as one, and Chaussier very appropri- ately had been given this chair in the Paris School of Health. His associate was Antoine Dubois, who was destined later to become one of the great deans of the Paris medical faculty. At noon on alternate days came lectures on medical chemistry and pharmacy by Deyeux (who seems to have had no associate). The remainder of the afternoon was devoted to practical exercises in the laboratories. In the summer semester, beginning the last of March, at 10 a.m. on odd days, came lectures in materia medica and botany by Pey- rilhe and Richard, and on even days, lectures in medical physics and hygiene by Halle and Pinel; with no definite assignments in the afternoons. Second-year students in the winter semester again listened to lectures and performed practical exercises in anatomy-physiology and medical chemistry and pharmacy, but at noon on alternate days there were in addition lectures in operative medicine by Sabatier and Boyer. In the summer semester the student again listened to lectures in materia medica and botany but added at noon daily lectures in pathology, external pathology on even days by Choppart and Percy, internal pathology on odd days by Doublet and Bourdier, the afternoons being devoted to a course in obstetrics by Alphonse Leroy and Baudeloque. Third-year students in the winter semester again attended lectmes in anat- omy, pathology, chemistry, pharmacy, and operative medicine, but they might J. M. D. Olmsted 467 devote their afternoons either to practical exercises or to study; in the summer semester they Hstened again in the mornings to the professors of materia medica and of pathology; and in the afternoons they followed the course in obstetrics on odd days only, and on even days at 4 p.m. attended lectures on legal medicine and the history of medicine by Lassus and Mahon. From the wording of the text it would seem that there might be a great deal of dupli- cation, since there is nothing to show, for example, that the lectures in anatomy-physiology occurring at the same hour for students in each of their three years were not the same in subject matter. If, however, the instructors rotated subjects so that they were different each year, it would seem that the instruction adapted for first-year students would not be suitable for third- years students, and vice versa. Perhaps duplication or even triplication was not considered undesirable. In any case, the inauguration of an official series of examinations, in which the various subjects were disposed of in a definite order, must have brought about a corresponding reorganization of the curriculum just nine years later. It w^as not long after the passing of the decree of 1794 establishing the three Schools of Health that Napoleon took over the reins of government and at the same time began his active military campaigns. The original provision for 300 students at Paris, 150 at Montpellier and 100 at Strasbourg did not suffice to supply his armies, and on May 1, 1802, a further decree under the Consulate permitted the creation of three additional "Schools of Medicine," as they were now called, "one of which must be especially devoted to the study of diseases of land and sea troops." It was not until June 9 of the following year that the provisions of this decree w^ere carried out in part by the creation of medical schools at Turin and Mayence. The year 1802 also saw the inaugura- tion of the system of competitive examinations for externships and internships so characteristic of modern French medical education. Arrangements were made for two such competitions yearly for externships, and a single one for the highest prize of a medical student's career, the internship. The first in- terns received their promotion September 13, 1802. Another consular decree regarding medical schools, of equal importance with the original Revolutionary decree of 1794, was passed on March 10, 1803. Its object was to require any person adopting the professions of medicine or surgery, or wishing to serve as a health officer, to undergo an examination and be duly accepted by examining boards under government control. Until now no diplomas had been issued by the medical schools established under the Revolution, but from this time on candidates who successfully passed their examinations were to bear through the authority of the Minister of the In- terior the title of doctor of medicine, or doctor of surgery, or health officer. Since all practitioners must hold a diploma, the claims of those who had obtained degrees from pre-Revolutionary Faculties of Medicine and Colleges of Surgery were recognized and provision was made for giving diplomas to those who had studied the art of healing between 1793 and 1803. Bichat, who 468 A French Medical Legacy had exercised a greater influence upon medicine than any other teacher, had died without either a degree or a diploma, and Paris was treated to the spec- tacle of the famous Dupuytren defending the thesis which he had written for his degree, and even the great Boyer himself, although nearly fifty at the time, had to submit to an examination by his less illustrious colleagues and pay for his diploma. The decree of 1803 also increased the length of the course in medical schools to a minimum of four years, established a system of five exami- nations (two of which had to be in Latin) and required the writing of a thesis (either in Latin or in French). The five examinations were to be passed in the following order: (1) anatomy and physiology; (2) pathology and nosology; (3) materia medica, chemistry, and pharmacy; (4) hygiene and legal medicine; (5) internal or external clinical practice, depending on whether the candidate wished the doctorate in medicine or surgery. Although the three schools had originally been founded to train health officers, this decree put the health officers in a separate and inferior class by themselves. In preparation for their examinations they might apprentice them- selves to a physician for six years, or follow hospital service for five years, or spend three consecutive years in one of the medical schools. The number of their examinations was limited to three: (1) anatomy; (2) elements of medi- cine; (3) surgery and ordinary practices of pharmacy. That the student was held to the rigid system of examination is shown in the old diplomas still in existence. We may take that of Francois Magendie as an example, since it has been examined by the present writer in the Musee Gilbert of the Paris Medical School where it was to be found, at least until 1939, along with other personal papers carefully preserved by Magendie. We find that he passed his examinations in "anatomy and physiology" on "10 frimaire an XIII" (December, 1804). The date for the passing of pathology and nosology is obscured by worm holes in the parchment, only the month, "messidor," being decipherable. Presumably the year was 1806, since among his papers is preserved a certificate exempting him from military service in 1805 on the grounds of "visceral obstruction," and there is no record of his attendance at the hospitals during that year. This makes it probable that he did not attend medical school during 1805 because of illness. Examinations in materia medica, chemistry, and pharmacy were completed "26 vendemiaire an XIV" (September, 1806). The date for examination in hygiene and legal medicine is no longer written according to the Revolutionary calendar, but reads "May 20, 1807." On February 22, 1808, he satisfied his examiners as to his proficiency in "internal clinics." His thesis having been approved and his fees paid, the diploma was awarded March 24, 1808. The final step in the evolution of the external forms of French medical education was taken by Napoleon in his imperial decree of March 17, 1808, which recreated the University of Paris and placed all public instruction throughout the Empire under its authority. This brought the present-day Paris Faculty of Medicine into existence as one of the five faculties of the J. M. D. Olmsted 469 University, the others being science, law, theology and, as a concession, arts. Article 12 of this decree states that "the existing five schools of medicine* shall form five faculties of the same name. They shall preserve the organization determined by the law of 19 ventose an XI" (March 10, 1803). The die of French medical schools had, therefore, already been cast in 1803, and with the awarding of the doctorate as the certificate of the attainment of a superior medical education beyond that of an officer of health, the modern scheme of organization of medical instruction had already been entered upon. The later decree merely brought the medical schools into the same system as other departments of public instruction, and the title of doctor of medicine or doctor of surgery was now conferred under the authority of the University. A study of the 1938-1939 announcement of courses given by the Faculty of Medicine at the Sorbonne (the last to come to the Library of the University of California) shows that there is still a close parallel between its provisions and the scheme outlined in 1803, Vv'hich was, after all, merely an extension of the plan proposed by Chaussier in 1794. Mornings for all students of medicine are to be spent in the hospitals. For the M.D. degree there are five sets of exami- nations in theoretical subjects, three clinical examinations, and a thesis. The five examinations in theoretical subjects come at the end of the different years in the following order: (1) mainly anatomy with some histology, physiology, medical physics, and medical chemistry; (2) mainly physiology with some anatomy, histology, medical physics, and chemistry; (3) pathology, experi- mental medicine, parasitology, and bacteriology; (4) obstetrics, medical and surgical pathology; (5) general pathology, hygiene, legal medicine, therapeu- tics, and pharmacology. Aside from the inclusion of those sciences which were developed chiefly by the aid of the microscope in the nineteenth century, such as histolfjgy and bacteriology, the scheme is essentially the same as the original one, although naturally the content of the various courses has kept pace with advance in the medical sciences. The idea is to produce practitioners, hence the emphasis on hospital train- ing from the first day the student enters medical school to his last. The bad features of the system have been emphasized in the report (1932) of the Com- mission on Medical Education under the chairmanship of A. Lawrence Lowell. A single quoted sentence will convey the main objection: "The students are preoccupied and quite overwhelmed with clinical problems, for the under- standing of which they have had no preparation." In spite of this overatten- tion to the clinic, the Commission claimed that because of the competitive appointments of externs and interns, "50 per cent of medical students in France do not receive adequate clinical experience before graduation. About 40 per cent (those who have succeeded in becoming externs) obtain a good preliminary experience . . . while about 10 per cent (the interns) receive what is probably the largest and most varied clinical experience provided for med- ical students anywhere in the world." * Paris, Montpellier. Strasbourg, ruriii. and Mainz. 470 A French Medical Legacy The sick in Paris may therefore bless or curse the legacy of the Revolution according to whether or not they are attended by one of the favored lo per cent— which seems an odd result of at least one of the three principles of Liberty, Equality, and Fraternity. Relevant Literature Busquet, P.: "Chaussier (Francois)," in Les Biographies Medicates, 1 (Paris: 1927), 37. Carpentier, A., and Du Saint, G. F.: Repertoire Generate Atphabetique du Droit Francois (Paris: 1896). De t'Enseignement Actiiel de ta Medecine et de ta Chirurgie (Paris: 1815?). Duvergier, J. B.: Cottection Comptete des Lois, Ordonnances, Regtemcns, Avis du Conseit d'Etat (Paris: 1836). Finat Report of the Coinniission on Medicat Education (New York: 1932). Guizot, F.: Essai sur I'Histoire et sur I'Etat Actuel de t'Instruction Publique en France (Paris: 1816). Memoire en Reponse a un Ecrit Anonyme, intitule Observations Presentees au Roi, sur la Facidte de Medecine (Paris: 1815?). Plan General de t'Enseignement dans I'Ecole de Sante de Paris. Imprime par ordrc du Comite d'Instruction Publique de la Convention Nationale (Paris: An III de la Republique). Rapport et Decret de la Convention Nationale, sur les Ecoles de Sante de Paris, MontpelUer et Strasbourg (Paris: An 3 de la Republique). Reflexions sur I'Etablissenicnt d'une Societe Royale de Medecine et de Chirurgie (Paris: 1815?). Richerand, A. B.: "Seance publique de I'Ecole de Medecine, le 24 nov., i8o8'" Gaz. natn., ou Le Monit. univ. p. 1431, Dec. 29, 1808. i M CYTOLOGICAL DIFFERENCES BETWEEN CASTRATION AND THYROIDECTOMY BASOPHILS IN THE RAT HYPOPHYSIS By J. D. REESE, A. A. KONEFF, AND P. WAINMAN From the I)I\ [SIGN OF ANATOMY, MEDICAL SCHOOL AND THE INSTITUTE OF EXPERIMENTAL BIOLOGY, UNIVERSITY OF CALIFORNIA BERKELEY, CALIFORNIA CYTOLOGICAL DIFFERENCES BETWEEN CASTRATION AND THYROIDECTOMY BASOPHILS IN THE RAT HYPOPHYSIS Introduction IN VIEW of the voluminous literature which has accumulated on the pituitary of the rat, it seems remarkable that there is no agreement on the identity or lack of identity of castration and thyroidectomy basophils. The question still exists whether they differ morphologically to such a degree that they may be considered different entities. Furthermore it is not settled whether these cells arise from the same or different stem cells or whether they are the same or different physiologically. In this paper it is our purpose to deal only with the morphological aspect of the problem. In a later commvmication, we shall state the results of experiments dealing with certain aspects of the physiology of these cells. Historical Review The effect of castration upon the anterior lobe of the rat has been discussed in such admirable treatises as those of Severinghaus.^"* Studies dealing with the anterior-lobe changes after thyroidectomy have also been reviewed in recent years by Severinghaus''^ and by Zeckwer and co-workers." Agreement exists regarding some points of similarity and difference in the histology of the anterior lobe after the two operations. The granulated acido- phils of the rat pituitary disappear after thyroidectomy, according to Hohlweg and Junkmann;'' Severinghaus;^ Zeckwer;" and Guyer and Claus.^ It is also true that, in the rat, after both thyroidectomy and castration, there is an increase in the number and size of the basophils. In both conditions, basophil vacuola- tion is sooner or later a prominent feature. That these changes in the baso- phil are consequences of castration was established by the Avork of Addison," Schleidt,'" Lehmann," Nukariya,"^ and others. That such transitions in baso- phils followed thyroidectomy was established by Kojima;" Hohlweg and Junkmann;' Severinghaus, Smelser and Clark;' Zeckwer and co-workers;" and Guyer and Glaus.* In 1934 Severinghaus, Smelser and Clark,^ describing the basophils after thyroidectomy, reported the following: "The basophils are increased in num- ber, are of maximum size, and give to the pituitary the castrate appearance. Many have deeply chromatic elongated nuclei. Large numbers of typical cas- tration cells are present." In several later publications Severinghaus has ex- pressed the opinion that, except for the rapidity Vvith which the changes in the basophils take place, they are essentially the same after castration and thyroid- ectomy. [47-):] 474 Castration and Thyroidectomy Basophils Zeckwer and her co-workers' state: "Morphologically there are distinct dif- ferences between the changes in the pituitary resulting from thyroidectomy and those resulting from castration. It is only at long intervals after thyroid- ectomy that the cells are somewhat similar to castration cells. The intracellular globules of 'castration cells' are large, smooth, well defined, and appear to be composed of dense secretion, hyaline in appearance when fixed and stained, and push to one side the basophilic granules in the cell, but in thyroidectomy cells the secretory accumulation has a less regular contour. We have not seen in early stages of 'castration cells' the characteristics observed in early stages of 'thryoidectomy cells.' In both cases, thyroidectomy and castration, there is preliminary to the formation of vacuoles, a distinct increase in large solid basophilic cells." Guyer and Claus^ agree there are significant morphological differences between castration and thyroidectomy basophils and point out a number of criteria which they believe serve to differentiate the two cell types. Materials and Methods The present studies on the pituitaries of castrated and thyroidectomized rats have extended over a period of years— since 1930. The material consists of pituitaries of adult male rats which were castrated and sacrificed at periods varying from 10 days to 18 months after operation; also the glands of similar animals thyroidectomized and autopsied 10 to 162 days after thyroidectomy. Completeness of thyroidectomy was checked at autopsy by examination of the region under the binocular microscope. The following technical methods were employed for this study of the pitui- tary: Mallory-Azan technique of staining was the routine method for general histological study; this method was supplemented by cytological techniques similar to those used by Severinghaus; the Nassonov-Kolatchev method of osmic-acid impregnation was used to demonstrate the Golgi apparatus; some of such osmicated preparations were counterstained with the Mallory-Azan technique; the Altmann-Masson method of staining, following Champy fixa- tion, has also proved very helpful in demonstrating the mitochondria. Details of the histological and cytological methods as adapted here have been pre- viously published— Koneff;" and Reese, Koneff and Akimoto.^ Observations Insufficient attention has been given to the importance of the differences in the manner in which vacuoles form in thyroidectomy and castration cells. It is true, fallacious conclusions may be drawn from constructing a series of changes based upon a study of various stages in a process. However, if the intracellular phenomena are so marked as to result in reduction of the entire cytoplasm to a vacuolated mass, it does seem that a trained observer should not encounter any difficulty in identifying the various stages in the process. This is particularly cogent since the examination of preparations made at Reese, Koiieff, and Wainman 475 varying postoperative times can be correlated with the changes in individual cells in preparations representing a single postoperative period. It has been pointed out by Guyer and Glaus/ by Zeckwcr and co-workers," and others that more vacuoles are present in thyroidectomy than in castration basophils. In our preparations the basophils seen soon after thyroidectomy show an enlargement of fine alveolar subdivisions which may normally be present in the cytoplasm. These multiple vacuoles in the process of develop- ment occupy a considerable part of the circumference of the cell. The altered areas of cytoplasm have a smooth hyaline appearance, their content is baso- philic. They are at first quite small (pi. i, figs, i and 2), vary in shape, being often polygonal, and are separated from one another by areas of cytoplasm, normal in appearance. The partitions at first may be of considerable thickness, but ultimately become mere strands between the enlarging vacuoles (pi. 1, figs. 4 and 5). Examination of a series of cells shows not only that these vactioles are formed first on the periphery of the cell, but that progressive vacuolation of the cytoplasm approaches more and more closely the nucleus as time goes on (pi. 1, figs. 4, 5, and 6). Accompanying this reduction in the amount of normal cytoplasm there is a redistribution of mitochondria, and certain changes in the position of the Golgi apparatus. Early in the process, certainly as soon as changes in the peripheral cytoplasm are discernible, the majority of the mitochondria become concentrated in the region surrounding the nucleus. They are to be found either in the zone of perinuclear cytoplasm or they may be strung out along the intervacuolar partitions where they may give the appearance of strings of beads. As in castration basophils they are never found in the vacuole itself. The Golgi apparatus assumes a position early in this process quite at variance with its usual situation in the cytoplasm. It is found in the nonvacuolated portion of the cytoplasm, hence near the nucleus (pi. 1, figs. 7 and 8). In more advanced stages of the thyroidectomy basophil evolution, the vacu- oles tend gradually to become fewer and larger. It is not an unwarranted assumption that this reduction in number and increase in size is due to coales- cence. This coalescence is usually noted first where the vacuoles are first formed, namely at the periphery of the cell. When coalescence of the vacuoles becomes moderately advanced, the zone of cytoplasm nucleus and Golgi ap- paratus remains connected to the periphery of the cell by strands of cytoplasm. Such partitions become progressively fewer and fewer in more advanced stages of the process. Often the entire cytoplasm of a thyroidectomy basophil is riddled by vacu- oles, except for a small portion containing the nucleus, Golgi apparatus, and mitochondria. This segment of cytoplasm extends from the region of the nucleus to the nearest adjacent portion of the cell membrane. Since the nu- cleus is typically eccentric this cytoplasmic mass may be quite small, and as the vacuoles coalesce, the granular material and the cell organelles come to 47^ Castration andThyroidectomy Basophils occupy a small area of cytoplasm at or near the cell membrane. In the end, one or two vacuoles may be present, and the nucleus, remains of Golgi ap- paratus, and mitochondria are at the periphery of the cell. The cell as a whole presents true degenerative chaijges. When this type of vacuole formation is compared with that seen in the so-called castration cell, it may be seen here also that in early stages the cyto- plasm has a finely alveolar appearance. Especially is this evident in Champy- fixed material. The vacuolation as it appears, however, has the form of a single (or at most double or triple) vacuole which has no perceptible relation to these cytoplasmic subdivisions. The vacuole appears as a small round or oval area of altered cytoplasm, hyaline in nature (pi. i, fig. lo). Its colloidal content stains less deeply than the surrounding cytoplasm, but is typically more basophilic than the colloidal material in the thyroidectomy cell. We concur in the opinion expressed by Severinghaus that vacuole formation in the castration cell is preceded by liquefaction of the granular material of the cytoplasm. The most outstanding characteristic of vacuole formation in the castration basophil is that the precursor of the vacuole is not a subdivision of the normal cytoplasm as it is in the case of the thyroidectomy basophil but begins as a separate vacuole in the cytoplasm which enlarges and distorts the cell. The gradual enlargement of the vacuole produces its effect upon the cyto- plasm itself and its contents. The nucleus usually retains its shape until the pressure of the vacuole forces it toward the cell membrane; it then accommo- dates itself to the remaining available space. It becomes elongated in many instances, and finally it may become intensely hyperchromatic and compact. The Golgi apparatus also adapts itself to the free cytoplasm, which in the castration cell forms a rim of varying width about the entire circumference of the cell. The Golgi apparatus, just as the nucleus, becomes elongated and fits itself between the vacuole and the cell membrane. It may lie close to the nucleus or may be separated from it by a considerable distance. Schleidt^" had a very good reason for applying the term "signet ring" to these cells. The signet part of the ring consists of that portion where nucleus and Golgi apparatus lie and the normal cytoplasm forms a continuous, albeit in highly distended cells very thin, rim about the entire circumference of the cell (pi. i, figs, ii, 12, 13, and 14). It is true that the Golgi apparatus and nucleus, in the majority of both thyroidectomy and castration basophils, come to occupy a position near the cell periphery, but, as has already been demonstrated, the processes which lead to this final disposition are entirely different in the two cases. It should be added that clear vesicles are to be found in the hyaline vacuolar material of both castration and thyroidectomy basophils. The size finally attained during vacuolation in thyroidectomy cells is greater than that in the "signet ring" cells of castration. Reese, Koneff, and Wainman 4^7 There are other difTexences in the cytoplasm distinguishable in the early stages of development which enable one to differentiate the two types of basophil. The cytoplasm of the thyroidectomy basophil is usually more baso- philic than that of the castration cell. The cytoplasm of these cells contains coarse floccular basophilic material which becomes more and more apparent as the vacuoles increase in size and is subsequently forced into a small space near the nucleus. The cytoplasm of young castration basophils, thotigh it may be alveolar in structure, is more uniform than the thyroidectomy basophil and is more finely granular. In the Nassonov-Kalotchev preparations, the cytoplasm of the cas- tration basophil is generally less osmiophilic than is that of the thyroidec- tomy cell. There are noteworthy changes in the Golgi apparatus in the two groups of basophils other than the changes in position already mentioned. These are the changes in size, shape, and internal structure of the Golgi apparatus. The Golgi apparatus is a satisfactory indicator in the normal rat hypophysis of the strain to which any individual cell belongs. Severinghaus' first stressed the im- portance of this fact, and results from this laboratory are in agreement with his findings. This does not, however, mean that the position of the Golgi apparatus has equivalent value as an indicator of cell type after physiological or pathological changes. Such conditions may alter the relation of the Golgi apparatus to the nucleus and even the finer morphology of the apparatus. Under such circumstances, if the Golgi apparatus only were used as the single criterion, great difficulty might be experienced in determining whether a cell belonged to the acidophil or basophil strain. In the thyroidectomy baso- phil, the Golgi apparatus is so altered that it may bear little resemblance to the classic description which Sever inghaus has given of that organelle in normal basophil. At least some of these modifications which the Golgi apparatus undergoes are coincident with changes in the cytoplasm, and doubt- less result from such changes. As the Golgi apparatus is crowded to^vard the nucleus, the originally rounded apparatus comes to lie as a cap against the nucleus (pi. i, figs. 7 and 8). That we are dealing here with a thyroidectomy basophil and not an enlarged acidophil is easily proved by the application of a suitable counterstain, such as the Mallory-Azan to positive impregnations of the Golgi apparatus. It will be remembered that in the castration basophil the relation of the Golgi apparatus to the nucleus is quite different from that in the thyroidec- tomy basophil. As Guyer and Glaus' noted, the Golgi apparatus not uncom- monly lies between the nucleus and the enlarging vacuole (pi. 1, fig. 13). This relationship is, however, far from invariable, the Golgi body very fre- quently coming to lie side by side with the nucleus, both bordering the vacuole (pi. 1, figs. 11, 12, and 14); in other words, the vacuole often begins in a portion of the cytoplasm quite distant from the nucleus and Golgi ap- 478 Castration andThyroidectomy Basophils paratus, and the enlargement of the vacuole probably determines the ultimate position of both of the other cell structures. The ease with which the Golgi apparatus is demonstrable in the thyroid- ectomy and castration cell differs markedly. Even a short time after thyroidec- tomy, the Golgi apparatus becomes more difficult to demonstrate than in the normal or castrate basophil. This is probably related to the increased osmio- philia of the cytoplasm of these cells. Early enlargement of the Golgi apparatus is seen in both the thyroidectomy and castration basophils. The shape and internal structure of the apparatus, however, is markedly different. The enlargement in the castration cell is such that it does not disturb the fundamental pattern of the double-walled sphere. It is true that at a time when the Golgi body has reached maximum size (some- what before vacuolation occurs) its outer surface may be somewhat crenated, but the whole structure remains rounded or oval (pi. i, figs. 15 and 16). However lightly such an apparatus is impregnated, it always has a sharp out- line and the component parts of its net are shown with clarity. The Golgi apparatus of the basophil also enlarges after thyroidectomy, but assumes peculiar shapes conforming to the contour of the limited cytoplasm in which it is forced by the multiple vacuoles. The apparatus in such cells has a very tenuous character; long filaments or strands are formed which are inter- rupted by areas which are relatively osmiophobic. Though the shape of the Golgi apparatus is distorted by pressure effects, it does not need to be con- sidered degenerate in early stages. Guyer and Claus^ have described the Golgi apparatus in thyroidectomy as "commonly granular or diffuse." This is true and this granularity or diffuse- ness is seen in many cells which show only early thyroidectomy changes as far as vacuole formation is concerned. Later during vacuolation there may be complete fragmentation and the remains of the Golgi body are then mani- fested only as a granular osmiophilic area. In some cells, parts of the apparatus may remain intact while the remainder is undergoing fragmentation. Disin- tegration of the apparatus is the rule in later periods after thyroidectomy. The enlargement of the apparatus in thyroidectomy cells is temporary; the final effect is destruction. Granular distintegration of the Golgi body is not seen in castration cells. In these cells the apparatus is first enlarged, later becomes smaller. The en- larging vacuole compresses the apparatus into a very small space at the cell periphery and the shape of the apparatus is changed, but the integrity of the net remains undisturbed. Even in the largest castration cells after the appara- tus itself is decreased in size there is no evidence of a dissolution such as is seen in thyroidectomy cells. The mitochondria also differentiate the processes occurring in the basophils after the two operations, both by their changes in size, shape, and distribution. The differences in these structures in the two cells are sufficiently marked so Reese. Koneff, and Wainman 4*79 that they are discernible, not only in preparations specifically stained, but in those stained by the Mallory-Azan technique in which the mitochondria are stained by azo-carmine. No experimental conditions have been encountered, aside from estriniza- tion, which produce mitochondria of such large size as have been observed in thyroidectomy basophils. These mitochondria are almost all oval or spherical in shape, filaments and rods being very rare. In almost 6 to lo per cent of the cells they are unusually coarse and numerous. The mitochondria are situated close to the nucleus or are strung like beads along the thin strands of cytoplasm which form the thin partitions between the numerous vacuoles in these cells (pi. i.fig. 9). After castration, the basophil mitochondria, while very numerous, are in general smaller, and are more dispersed throughout the cell; short rods and even filaments are present as well as granules. During the period of enlarge- ment of the castration basophil, the mitochondrial distribution usually fol- lows the normal pattern. They are definitely concentrated in the region of the Golgi apparatus (pi. i, figs. 15 and 17), from which they radiate into the surrounding cytoplasm. It is true that this distribution is disturbed by the appearance of the vacuole, but the fundamental arrangement of mitochon- dria, as ^vell as other intracellular inclusions, is not markedly affected until late in the process. A comparison such as has been undertaken here would be incomplete with- out some statement regarding differences in general cell size and shape at vai-ious periods after the two types of operation. These criteria alone allow the observer to differentiate at a glance the earlier phases of the two processes. Zeckwer and co-workers' have stated that in both instances there is "prelimi- nary to the formation of vacuoles a distinct increase in large solid basophilic cells." The contour, however, of these large basophilic cells differs remarkably and assists in the differentiation. In discussing this point it is necessary to stress the difference between the basophils of normal female and male rats. By this criterion we believe one can differentiate pituitaries from the two sexes. The predominant type of basophil in the normal pituitary of the male is an oval or rounded cell which stains comparatively lightly; in the female the predominating basophil is polygonal in shape and more basophilic, and in this cell the mitochondria are large and usually spherical. In the early response to thyroidectomy of either sex, a marked increase in number of basophils occurs, the cells closely approximating the type of cell just described for the normal female. They are irregular in shape, often polyhedral and have a dis- tinctly increased basophilia. As Guyer and Claus^ have stated, thyroidectomy basophils show a marked tendency toward grouping. Castration basophils, on the other hand, in early stages of formation are compact cells which are oval or round. The granules are evenly distributed and the cell stains as a rule more lightly with aniline blue than young thyroidectomy basophils. In a ^8o Castration andThyroidectomy Basophils pituitary studied about thirty-five days after castration, these cells are found to be exceedingly numerous and are scattered throughout the anterior lobe. They show no tendency toward grouping as after thyroidectomy. There are similarities between the late thyroidectomy and castration baso- phils, but these similarities are superficial. In the first place, the castration basophil distended by a castration vacuole never reaches the size characteristic of the basophil after thyroidectomy. The other criteria which differentiate the cells in these late stages are as follows: Castration basophils undergoing vacuolation are always accompanied by a crop of young basophils. These young basophils have been described as typical of the early castration reaction, but they are always present in appreciable numbers regardless of the period which may elapse after castration. Development of new basophils in such numbers does not take place after thyroidectomy. In fact after a relatively long postoperative period (62 days) very few young basophils are present; practically all the basophils show some degree of vacuolation. After 165 days all recog- nizable basophils are markedly diminished in number. The vacuolated cells by this time are sometimes recognizable only as irregular masses of basophilic cytoplasm and one must trace these in serial sections in order to identify the remains of the nucleus. The young basophils present are negligible in number and there is no evidence that the gland is making an effort to replace the vacuolated basophils. This diminution in the total number of basophils is therefore undoubtedly due to two factors— first, the vacuolated cells undergo complete degeneration, and secondly, a new crop of the basophils does not develop from the stem cell. Certain differences in the nuclei also serve to differentiate the basophils after thyroidectomy and castration. Guyer and Claus® have made the point that vacuolation of the nucleus is common in basophils after thyroidectomy, but does not occur in the same cells after castration. That this cannot be con- sidered an important differential point will be seen from observations on the nucleus changes. In the period following thyroidectomy, from 10 to 63 days, the following changes in the nucleus have been observed: Prior to the develop- ment of vacuolation, the nuclei of the polyhedral darkly staining basophils are often compact and polychromatic. In the initial periods of vacuole devel- opment the nucleus of the enlarged thyroidectomy basophil is round, lightly staining and vesicular; there is a notable scarcity of basichromatin. The size of the nuclei is often greatly increased, in fact may reach twice the size of the nucleus of the normal basophil. The nucleoli also are greatly enlarged, some- times threefold the size of those in the normal basophil (pi. 1, fig. 2). Later the nucleus shows definite evidence of degeneration and finally becomes pyknotic. This process continues until the nucleus is an irregular compact mass lying at or near the periphery of the cell (pi. i, fig. 6). Changes in the nucleus of the basophils after castration are different. Often one sees elongation of the nucleus, due doubtless to the pressure exerted by Reese, Koneff, and Wainman 48 1 the enlarging vacuole. Such a change in the contour of the nucleus is rare in the thyroidectomy basophil. On the other hand, enlargement of the nucleus and nucleolus described in the basophil after thyroidectomy are very un- common for the cells affected by castration. Although pyknosis may be the ultimate fate of the nucleus in both cell types, it is not so common a phenome- non after castration as after thyroidectomy. No difficulty was encountered in detecting the vacuolation of the nucleus found by Guyer and Glaus* in the basophils after thyroidectomy. These single vacuoles contained a homogeneous colloidlike material reacting like mucin with methyl green.* Later such vacuoles sometimes had increased in size so that they filled the nucleus. The nucleoli in such cases were pushed toward the nuclear membrane (pi. i, figs. 18 and 19). In the colloidal material of the nuclear vacuole, round or oval colorless spaces were sometimes present, pos- sibly due in part to solution of the material in reagents used in histological preparation. The growth of such intranuclear vacuoles may proceed to the point at which the vacuole itself is larger than the entire nucleus of a normal basophil, the nuclear membrane and nucleoli still being intact. Although vacuolation of the nucleus of the basophil is less common after castration than after thyroidectomy, it has nevertheless been observed in certain pitui- taries from rats sacrificed at long periods after castration. These vacuolar changes did not differ in any essential way from those seen after thyroidec- tomy (pi. 1, fig. 20). Vacuolation of the nucleus cannot therefore be considered to constitute a reliable criterion for differentiating the two types of basophil (as suggested by Guyer and Glaus*). Nevertheless there are differences in the nuclei which aid in the differentiation. The increase in size of the nucleus after thyroidec- tomy is more marked than after castration. The pressure of the castration-cell vacuoles often causes an elongation of the nucleus which is highly character- istic. Finally, the marked enlargement of the nucleoli seen after thyroidectomy is not present after castration. In discussing differences or similarities of castration and thyroidectomy basophils one should always bear in mind that "castrationlike" or "thyroidec- tomylike" basophils may occasionally be found in other types of hypophyses, including normal ones. The authors do not consider, however, that cells identical with those of castration occiu~ in significant numbers after removal of the thyroid. The crux of the entire problem is whether any competent observer can distinguish the pituitary of a thyroidectomized animal from one which has been castrated if the decision is based only upon changes in the basophils. Using the criteria described in this paper it is considered that one should not encounter difficulty in distinguishing the two types of pituitaries regardless of the postoperative period. * Unpublished data of A. KonefT show the effecti\eness of methyl green in staining mucoid substances. 482 Castration andThyroidectomy Basophils Conclusions The thyroidectomy and castration basophils of the anterior hypophyses of the rat differ morphologically to such an extent that they may be used to differentiate the pituitaries after these two types of operation. These differ- ences are manifested in the size, shape, and distribution of the cells, also in the mode of vacuole formation and in the distribution and fate of cytoplasm, Golgi apparatus, and mitochondria. REFERENCES 1. Severinghaus, A. E.: Anat. Recrd. 57:149, 1933- 2. Severinghaus, A. E.: Proc. Assn. Resrch. Nerv. & Ment. Disease 17:69, 1936. 3. Severinghaus, A. E.: Physiol. Revs. 17:556, 1937. 4. Severinghaus, A. E.: in Sex and Internal Secretions, ed. by E. Allen (2d ed.; Baltimore: 1939). 1045- 5. Severinghaus, A. E.; Smelser, G. K., and Clark, H. M.: Proc. See. Exper. Biol. & Med. 31: 1127, 1934. 6. Zeckwer, I. T.; Davidson, L. W.; Thomas, B. K., and Livingood, C. S.: Amer. Jl. Med. Scis. 190:145' 1935- 7. Hohlweg, W., and Junkmann, K.: Pfliigers Arch. f. d. gsmt. Physiol. 232:148, 1933. 8. Guyer, M. F., and Glaus, P. E.: Anat. Recrd. 67:145, 1937. 9. Addison, W. H. F.: Jl. Compar. Neurol. 28:441, 1917. 10. Schleidt, J.: Zentrlblt. f. Physiol. 27:1170, 1914. 11. Lehmann, J.: Pfliigers Arch. f. d. gsmt. Physiol. 216:729, 1927. 12. Nukariya, S.: Pfliigers Arch. f. d. gsmt. Physiol. 214:697, 1926. 13. Kojima, M.: Quart. Jl. Exper. Physiol. 11:319, 1917. 14. Koneff, A. A.: Stain Technol. 13:49, 1938. 15. Reese, J. D.; Koneff, A. A., and Akimoto, M. B.: Anat. Recrd. 75:373, 1939. EXPLANATION OF PLATE (Reduced 1/6) PLATE 1 All photographs were taken with a 2 mm. oil immersion objective; magnifica- tion X 900. Figs. 1, 2, 4, 5, 9, 17 were taken from Altmann-Masson preparations; figs. 3, 7, 8, 12, 13, 16 from Nassonov-Kolatchev preparations; figs.'G, 10, 11, 14, 15, 18, 19, 20 from section stained by Mallory-Azan method. Figs. 1 to 6 represent the sequence of changes observed in basophils after thyroidectomy. Figs. 10 to 14 represent the sequence of changes observed in basophils after castration. Figs. 7 and 8. Highly reticular and diffuse Golgi apparatus found in the baso- phil of thyroidectomy. Fig. 9. Thyroidectomy basophil. Note the distribution of mitochondria in the remaining part of the cytoplasm near the nucleus and between vacuoles. Figs. 15 and 16. Negative image and a positively impregnated Golgi apparatus of the yoimg castration basophil without the vacuole. Fig. 17. Note the general distribution and concentration of mitochondria in the Golgi region of a young castration basophil. Figs. 18 and 19. Small early and achanced nuclear vacuolation found in the basophil after thyroidectomy. Fig. 20. Castration basophil sho\\'ing vacuolation of the nucleus morphologi- cally identical with that seen in the thyroidectomy basophil (fig. 18). [ 484 1 STUDIES ON THE GROWTH OF LYMPH NODES, THYMUS, AND SPLEEN IN THE RAT By WILLIAM O. REINHARDT From the DIVISION OF ANATOMY AND THE INSTITUTE OF EXPERIMENTAL BIOLOGY, UNIVERSITY OF CALIFORNIA BERKELEY, CALIFORNIA STUDIES ON THE GROWTH OF LYMPH NODES, THYMUS, AND SPLEEN IN THE RAT Introduction EXPERIMENTAL and clinical observations on changes in size of the lymph nodes, thymus, and spleen in various pathological conditions have sug- gested the possibility of making comparative quantitative studies on the growth and reaction of these structures to disturbances of an endocrine or nutritional nature. In order to lay a foundation for studies of a quantitative natine on lymph nodes, spleen, and thymus, it was thought advisable to follow the development of these organs over a considerable part of the life span of the rat in order that their behavior in disturbed physiological states might be inore intelligently interpreted. The form of the growth curves for thymus and spleen have been studied elsewhere in other strains of rats than the Long-Evans strain (Donald- son,^ Chiodi,^'^ Plagge*). The present study is devoted to a comparison of the growth in weight of thymus and spleen with that of the lymph nodes in the Long-Evans strain. The growth in weight of these organs is followed in groups of rats born at approximately the same time and autopsied at intervals over a period of seven months. In addition, the reactions (in terms of weight changes) of these structures to such stimuli as fasting, exposure to low temperature, and endocrine imbalances are indicated to show the extent of weight loss which may occtu^ in these structures. Histological studies are not included in this report. The studies of Job' have been of value in elucidating the location of the various grotq^s of lymph nodes in the rat. It was demonstrated by Job, and is confirmed by the present worker, that the lymph nodes in the rat are relatively constant in position and number. On the basis of unpublished experiments, the cervical and mesenteric nodes have been selected as representative of the lymph node tissue in the body. It is, of course, realized that the behavior of these groups of nodes may not be representative of all lymjDh nodes as a whole; generalizations may be applied only in a limited way to the other lymph nodes of the body. Materials and Methods Rats of the Long-Evans strain born in the last week of June and the first week of July, 1941, were maintained under comparable environmental conditions to the age of seven months. The animals received a regtdar ration (Diet XIV) ad libitum, supplemented twice weekly with fresh lettuce. At varying intervals during this time, groups of each sex were weighed and sacrificed under chloro- form anesthesia followed by bleeding from the heart. Animals from the same litters were sacrificed in different age groups to avoid peculiarities due to litter groupings. The age at autopsy and the number of animals in each group are indicated in the tables. The animals were then dissected, the tissues being 1:489: ^go Lymph Nodes, Thymus, and Spleen removed in a definite order by the same person to avoid variation in technique. Upon removal from the animal, the tissues were placed in moist-atmosphere weighing dishes and were weighed to the nearest milligram in a closed balance. Care was taken to avoid drying of the tissues before and during weighing. The technique remained the same for the entire series of autopsies. The data ob- tained were analyzed to show means and standard errors of the means. These data are presented in the form of tables, and further in the form of curves, showing changes in organ weight for each age group. The body-weight aver- ages included in the tables are not corrected for intestinal and bladder con- tents, nor were the animals fasted preliminary to autopsy. A brief description is in order as to the technique of dissection of spleen, thymus, and lymph nodes. The spleen was removed immediately after the death of the animal in order to eliminate variation in weight due to changes in the blood content of this organ. The spleen and thymus were dissected out by simple severance of their connective tissue and vascular pedicles. Care was taken in the case of the thymus to dissect away the associated lymph nodes. Dissection of the cervical lymph nodes was carried out in the following man- ner: a midline incision from the tip of the mandible to the sternum was car- ried down through skin and superficial connective tissue. Reflection of the skin flaps allowed exposure of the lymph nodes where they lie lateral to the submandibular salivary glands along the line of the external jugular vein. Incision of the overlying connective tissue allowed the enucleation of the well- encapsulated nodes. The pedicles of the nodes were severed along the line of the external jugular vein and the nodes were cleanly excised. There were usually three separate and discrete nodes in the position mentioned. Included in the nodes dissected was one deep cervical node from each side of the neck (located dorsal and lateral to the omohyoid muscle where the latter crosses the vascular sheath of the neck). The mesenteric nodes required relatively more care in dissection. The procedure employed was to remove the entire gastroenteric tract and associated viscera from the abdomen by simple section of the mesenteries at the points of reflection upon the abdominal wall. The liver was removed from the excised group of organs, and the entire mesentery was carefully stripped away from the stomach and the large and small intes- tines. Remaining was the mesentery (of the intestines and stomach) with the associated vessels, fat, nerves, and lymph nodes. The mesenteric nodes were found lying in a row along the root of the mesentery. By simple incision of the overlying peritoneum the nodes were removed and freed from connective tissue and fat. The nodes of the cervical and mesenteric regions were solid structures, the cut surfaces appearing rather dry, there being little fluid accumulated within the nodes. The dissection of the mesenteric nodes included groups which may be designated as hepatic, gastric, intestinal, pancreatic, and cecal. The animals autopsied were examined for the presence of pathological con- ditions. Animals suffering from gross evidence of respiratory or enteric dis- turbances were excluded from the study. William O. Reinhardt 491 Results The data obtained in this study of the growth of thymus, spleen, and lymph nodes are summarized in tables 1 and 2. A total of 284 animals were em- ployed, divided equally between males and females. Seven age groups were studied, consisting of about 20 animals of each sex. The lymph nodes were not dissected in the newborn animals because of practical difficulties. Study of TABLE 1 Mean Weights of Thymus, Lymph Nodes, and Spleen of Male Rats Age in days No. of animals Body weight, gm. Thymus Spleen Cervical nodes Mesenteric nodes I 20 8 5± 0. 2 ± 1-3 II. 5 ± 0.5* 128 ± 7.5 33 -S^ 2 224 it 12 22 22 47 25± 1.5* 46it 2.2* 42 21 128 ± 3 356 ±19 593 ± 56 128 1 12 228±17 65 16 ■"-33 ± 8 342 ±20 1135 ±122 2i8±i4 295±I4 100 18 306 ±10 260 ± 1 3 1350 ± 56 35i±i8 296±I2 176 20 394 ±15 191 ±18 1203 ± 48 257±i5 289±i5 223 20 430 ±11 171 ±12 1204 ± S3 275±i6 299±i5 * Mean weights in milligramsistandard error of the mea integer. "(^)- All figures expressed to the nearest TABLE 2 Mean Weights of Thymus, Lymph Nodes, and Spleen of Female Rats Age in days No. of animals Body weight, gra. Thymus Spleen Cervical nodes Mesenteric nodes I 25 6.7ibO. 2 8.9± 0.4* i9-7± 1-3 21 20 49 ±1-5 iqo ±12 334 ±16 29± 1.4 57± 3-7 42 21 115 ±3 308 ±13 625 ±73 I25±IO 2I5±I4 70 21 166 ±4 306 it 13 832 ±72 igiztii 265± 9 95 20 210 =h5 235 ±14 878 ±64 27i±i6 273±ii 176 20 245 ±8 156 ±10 781 ±53 212±15 275±i5 223 20 254 ±6 142 it 9 747 ±45 i66± 9 244±i4 * As in Table i. the standard errors in relation to the number of samples and the size of the means provides an estimate of the amount of variation in these tissues. Such analyses, however, are not to be considered as standards in application to other experiments. They are included here for the purpose of indicating the extent of observed variation in this series of animals and under the conditions employed. Figure 1 shows the growth in weight of the thymus gland in both males and females in this series of animals. Each of the points on the curves represents, on the average, the mean of the thymus weights of 20 animals at the various ages indicated. 492 Lymph Nodes, Thymus, and Spleen Examination of these curves shows the absolute-weight maximum of the thymus to occur between the ages of 40 to 60 days. This is supported by studies of the thymus weight of some six hundred other animals of various ages in this colony. After the attainment of maximum absolute-weight there is a ■100 Ceri/ica/ Lymph Nodes /Z5- /50 //s 200 22S Aae in Days Fig. 1. Growth in weight of thymus, spleen, and lymph nodes in male and female rats. gradual decline. Studies of the relative-weight curves (not included here) show the maximum relative weight to be attained between the ages of 25 to 45 days. The relative weight is high at birth, rises rapidly, and falls off proportionately as involution of the gland gains momentum. On the basis of these studies, one may consider the thymus as the most labile of the structures studied in that its growth and involution are the most pronounced. The form of the curve ob- tained for the growth of the thymus is similar to that obtained for other colonies, differing chiefly in the time of attainment of maximum absolute- William O, Reinhardt ^g^ weight. This may be related to colony differences, environmental conditions, or differences in the onset of puberty. Figure i also presents the curves for the growth in absolute weight of the spleen. No correction has been made in the means for the variation in weight of the spleen which occurs within the groups and which is evidenced by the large standard errors. It is realized that weight studies of the spleen must take into account the relation between the blood and lymphoid elements of that organ. An attempt was made to standardize the technique of autopsy and dis- section in order to render this relationship as nearly constant as possible, although other factors may complicate this problem. The average values for the weight of the spleen are greater for the male than for the female after the age of 50 days. There is a gradual leveling off in the absolute weight of the spleen in each sex, but studies of the relative weights show that there is a com- parable decline for each sex after 100 days of age. The reason for the greater absolute weight of the spleen in the male is not apparent. Donaldson^ states that there was no apparent difference between male and female spleen Aveights in material employed by him in his studies. The form of the growth curve for the cervical lymph nodes is similar in males and females. There is an increase in weight to the age of 100 days and then a gradual decline, but not as marked as in the case of the thymus. A study of relative-weight curves shows the same phenomenon as is the case for the absolute weights. One may speak of growth and involution in weight of the cervical lymph nodes on the basis of these curves. The form of growth curves for the cervical lymph nodes is generally similar to those of Hellman* for the rabbit, and those of Chiodi* for the rat. These investigators employed the same technique as here utilized, but they have examined other groups of nodes in smaller numbers of animals. Growth of the mesenteric lymph nodes in absolute values is the same for males and females. Maximum weights in both sexes are attained at about the age of 60 days, after which time the weights remain fairly constant. Relative to body weight, the weights of the mesenteric lymph nodes show a slow decline (after the age of 40 days). Since the absolute weights of the mesenteric nodes remain constant for the various age periods after the age of 60 days, there is relatively more mesenteric node tissue in the female because the latter does not attain the same body weight as the male. This contrasts with the case of the cervical lymph nodes in which the form of the absolute- and relative-weight curves for male and female are more nearly the same. Since the mesenteric nodes apparently have a different function from that of the cervical nodes, particularly in relation to the absorption of food from the intestine, it is pos- sible that the growth and maintenance of the former may depend on other factors, such as food absorption. Aside from interest in a comparison of the growth curves of the thymus, spleen, and lymph nodes, the value of these studies lies in their application to experiments devoted to elucidation of the function of these structures in the 494 Lymph Nodes, Thymus, and Spleen animal economy. The relationship may be made clearer by comparative studies of the effects, on lymph nodes and spleen, of endocrine imbalances and dietary disturbances which are already known to affect the thymus gland. The Effect of Fasting and Exposure to Cold on the Weights of Thymus, Lymph Nodes, and Spleen The following experiments are included to show the extent to which disturbed nutrition will influence the weights of the structures under consideration. Fasting is well known as a stimulus which will cause rapid weight involution TABLE 3 Effect of Fasting on the Weights of Thymus, Lymph Nodes and Spleen IN Male Rats Treatment No. of animals Final Body weight, gm. Thymus Lymph nodes Mesenteric nodes Spleen Fed controls 3 317 (210-332) 306* (292-320) ' 453' (329-568) 328 (263-404) 695 (560-862) Fasted controls 3 179 (162-204) 66 (36-95) 224 (142-267) 89 (46-117) 301 (188-381) Fasted 5 days, refed 3 days 3 243 (204-305) 137 (119-171) 351 (308-405) 4IO (234-678) 741 (576-923) * Weight of the tissues in milligrams. Figures in parentheses indicate ranges. of the thymus gland. It was thought probable that the lymph nodes might be affected in a similar manner. Normal male rats of this colony, 80 days of age, were subjected to fasting for a period of five days. The animals had access only to tap water for drinking purposes. At the end of this time, normal control fed animals of the same age were autopsied at the same time as the experimental animals. A further group was fasted for five days and then refed the regular diet for three days. Data are given in table 3. The findings have been confirmed in subsequent experiments. The lymph nodes dissected included the com- bined cervical, axillary, mediastinal, and retroperitoneal abdominal nodes, and the mesenteric nodes which are noted separately. Thymus, lymph nodes, and spleen, as can be seen by examination of this table, undergo a marked diminution in weight subsequent to a period of fasting. Refeeding restores these structures to an approximately normal weight even though the animals have not regained their original body weight. This contributes direct evidence that not only thymus but also lymph nodes and spleen respond to the withdrawal of food by marked decrease in weight. A further experiment to demonstrate the gross reaction of these organs to sudden and profound changes in environment was carried out by exposure of William O. Reinhardt 495 animals to a cold atmosphere. Female rats averaging 210 gm. in body weight were subjected to exposure in the cold room for a period of 63 hours to a tem- perature of 36-38° F. and a further exposure to a temperature of 25° F. for a period of three hours. The animals were placed in individual wire-bottom metabolism cages and allowed free access to 1 per cent sodium chloride solu- tion as a source of drinking water. The groups employed consisted of normal fasted animals maintained at room temperature, a group exposed to cold and fasted, and a third group which was allowed food ad libitum in the cold. The TABLE 4 Weights of Thymus, Spleen, and Lymph Nodes in Rats Exposed to the Cold Treatment No. of animals Thymus Spleen Cervical nodes Mesenteric nodes Normal fasted 4 178* (134-262) 804 (660-1089) 234 (164-368) 325 (278-391) Exposed to cold and fasted 5 89 (61-II7) 363 (202-583) 116 (72-164) 126 (105-150) Exposed to cold and fed 4 144 (102-190) 472 (394-615) 182 (150-231) 228 (198-263) * Organ weights in milligrams. Figures in parentheses represent ranges within each group. thymus, spleen, cervical, and mesenteric nodes were weighed. The data ob- tained are presented in table 4. It would seem reasonable and in agreement with Selye' to conclude that ex- posure to cold constitutes a stimulus which will cause a significant absolute- weight reduction in the organs studied and that this weight reduction may be partially prevented by the administration of food. Relationships between Endocrines and the Thymus, Spleen, AND Lymph Nodes Experiments involving the effect of certain hormonal treatments or deficien- cies on the thymus have in turn led to a consideration of the role of certain endocrine glands in affecting the gross amount of lymph-node tissue in the body. In this laboratory, studies have been made of the effect of adrenalectomy, castration, and hypophysectomy. The adrenalectomized animal maintained on 1 per cent NaCl solution has shown significant increases in certain of the lymph-node groups (Reinhardt and Holmes*). Castration causes a significant enlargement of thymus, spleen, and lymph nodes (Reinhardt and Wainman"). These findings are in accord with those of Chiodi' and of Houssay and co- workers.^" Preliminary studies show that hypophysectomy causes a decrease in the weight of thymus, spleen, and cervical lymph nodes, but not of the mesen- teric lymph nodes. Administration of desiccated thyroid gland orally to nor- 496 Lymph Nodes, Thymus, and Spleen mal rats causes a marked weight increase in lymph nodes and spleen, but not in the thymus. Testosterone propionate and estradiol dipropionate have only moderate reducing effects on the weight of the lymph nodes as contrasted with the usual marked involuting effects on the thymus gland. Interesting is the fact, however, that testosterone propionate will prevent the increase in the weight of lymph nodes which follows castration of the male rat (Reinhardt and Wainman"). Other preliminary experiments indicate that anterior pitui- tary growth hormone will cause increase in weight of thymus and lymph nodes in both normal and hypophysectomized animals, but will not produce en- largement of lymph nodes of the hypophysectomized rat when the food intake is restricted to the level of untreated hypophysectomized animals. This places further emphasis on the role of nutritive substances in the maintenance of the normal amounts of lymph-node tissue in the body. Further experiments are planned to show the role of certain vitamins in this respect. Summary (i) NovTiial Growth and hivohition. Studies of the weights of thymus, spleen, cervical and mesenteric lymph nodes in a series of 284 rats of both sexes and of various ages (birth to 7 months) have been used as the basis for the con- struction and comparison of the curves for growth in weight of these- organs. The cervical and mesenteric nodes have a period of growth and involution in weight which is comparable to that of the thymus and spleen with certain exceptions which are discussed. The absolute-growth curves for male and female animals for the thymus, cervical and mesenteric nodes are generally the same in form. The maximum absolute weight of the thymus is reached at about the age of 40 days, of the spleen, cervical and mesenteric nodes be- tween the ages of 60-100 days. After the attainment of maximum absolute weight there is a decline in weight which is most rapid for the thymus and least for the mesenteric nodes. Maximum relative weights are attained for the thymus, spleen, and mesenteric nodes at about 35 days of age (25-50), ^vhereas the cervical nodes attain their maximum at about 100 days. There is a rapid decline in the relative weight of the thymus and a less rapid decline in the relative weights of the spleen, cervical and mesenteric lymph nodes. One may speak of a weight increase and a weight involution, of a greater or lesser degree, depending on the organ concerned, over the period of the life span of the rat here studied. (2) Reaction of Lymph Nodes, Thymus, and Spleen to Fasting, and Expo- sure to Cold and Certain Hormonal Stimuli. Employment of quantitative measurements of changes in weight of the above structures demonstrates that the rapid weight involution which characterizes the reaction of the thymus to fasting and cold is also found to occur in the case of the spleen and lymph nodes. Certain hormonal stimuli which are known to enlarge or decrease the size of the thymus are discussed as exerting a generally comparable effect on lymph nodes and spleen. William O. Reinhardt 49'7 REFERENCES 1. Donaldson, H. H.: Mems. Wistar Inst. Anat. & Biol. no. 6, 1924. 2. Chiodi, H.: El Timo en Relacion con el Crecimiento y la Funcion Sexual (Thesis; Buenos Aires: 1938). 3. Chiodi, H.: Rev. Soc. argentin. de biol. 14:74, 1938. 4. Plagge, J. C: Jl. Morphol. 68:519, 1941. 5. Job, T.: Anat. Record. 9:447, 1915- 6. Hellman, T. J.: Upsala lakareforengs. forhdlr. (n.f.) 19 (supp.), 1914. 7. Selye, H.: Endocrinology 21:169, 1937. 8. Reinhardt, W. O., and Holmes, R. O.: Proc. Soc. Exper. Biol. & Med. 45:267, 1940. 9. Reinhardt, W. O., and Wainman, P.: Proc. Soc. Exper. Biol. & Med. 49:257, 1942. 10. Houssav, B. A.; del Castillo, E. B;, and Pinto, A.: Rev. Soc. argentin. de biol. 17:26, 1941. THE SELF-SELECTION OF DIETS By CURT P. RICHTER From the PSYCHOBIOLOGICAL LABORATORY, PHIPPS CLINIC JOHNS HOPKINS HOSPITAL BALTIMORE, MARYLAND THE SELF-SELECTION OF DIETS Di'RiiXG the last three decades workers in agricidtural stations and in nutri- ' tion laboratories have made sporadic attempts to test the ability of ani- mals to select their own diets. Eward^" reported the first observations on farm animals. He found that hogs which had access to a number of different food- stuffs—whole meal, meat meal (60 per cent protein), whole oats, linseed-oil meal, wheat middlings, charcoal, limestone, salt, water, all offered in separate containers— showed excellent growth. One hog was the largest that had been raised at the Iowa Agricultural Station. Pearl and Fairchild^ reported that chickens which had access to a variety of natural foods grew better than those that were given the regular diet. Dove^ also found that chickens thrived on self-selection diets. Lewis," and Stearns and Hollander" successfidly used "cafe- teria" feeding systems for pigeons. The latter authors found that choices made from corn, peas, wheat, and kaffir corn, offered in separate compartments, greatly speeded up egg production and spillage and wastage of food were re- duced. Godden" found that animals grazing on uncultivated pastures selected grass from parts of the pastures which had a higher mineral content. Orr^ has reported numerous instances of self-selection of diets in cattle. One of the most common observations is that animals that graze in pastures with a low phosphorus content manifest a craving for foods with a high phosphorus content. The observations made by Theiler, Green and Viljoen" on cattle in South Africa constitute a classical example of reactions of animals to a phos- phorus-deficient diet. Osborn and Mender" made the first self-selection studies on rats under laboratory conditions. They found that when rats were offered pairs of diets, one inferior, the other superior, the rats regularly selected the latter. Mitchell and Mendel" later made similar observations. Studies have also been made on human beings, most notably by Davis,^'^^ who foimd that very young children given a free choice of a great variety of natural foods grew and thrived. Sometimes for several days a child ate scarcely anything but eggs, or large amounts of fats, and the intake of the different foods fluctuated markedly from day to day, or week to week. Sweet" has also reported favorable results of free-choice methods of feeding children. In most instances workers interested in self-selection studies have based their studies on the belief that animals or human beings could make beneficial selections only from so-called natural foods, that is, foods which were natural to animals of a given species. More recent experiments, in which it was shown that rats are able to make beneficial selections from purified substances, have opened up a new attack on the problems concerned with the self-selection of diets. From these studies, made chiefly on rats, we have learned the following principles: i.Nor7nal rats make beneficial selections from purified substances: carbo- hydrates, fats, proteins, minerals, vitamins. Rats have been kept several hundred days on a choice of sixteen purified substances offered in separate to2 Sel£-Selectioii of Diets containers, using dextrose or sucrose as a source of carbohydrate, casein (autoclaved and purified) for protein, and olive oil for fat; crystalline thiamin chloride, riboflavin, nicotinic acid, pyridoxin, pantothenic acid, and choline chloride as representatives of the vitamin B complex; cod liver oil for vitamins A and D; and solutions of sodium phosphate, sodium chloride, calcium lactate, magnesium chloride, and potassium chloride for the minerals. Some of these rats gi-ew at a normal rate and showed regular vaginal smear cycles (Richter and Hawkes^^. 2. Rats kept on a vitamin-deficient diet will seek vitamins or substances which help to replace vitamins. Thus, rats kept on this self-selection diet with- out any of the vitamin B factors will eat large amounts of yeast or liver when these substances are offered to them and as a result quickly correct their deficiency. If not given access to a source of the B complex the rats will eat almost no carbohydrate, but take large amounts of fat, which apparently can be at least partially metabolized under these circumstances (Richter, Holt, Barelare and Hawkes^*). 3. The appetite for carbohydrate and fat always shows an inverse relation- ship. When rats have a marked craving for fat, they refuse carbohydrate and vice versa. Many instances of this relationship have been observed. Thus, vitamin B-deficient rats stop eating carbohydrate and eat large amounts of fats. Almost at once after they have access to vitamin B in the form of yeast, they reverse their appetites, stop eating fat and eat large amounts of carbo- hydrate. Rats made diabetic by pancreatectomy stop eating carbohydrate and start eating fat, then when treated with insulin reverse their appetite. After ligature of the bile duct, rats stop eating fat and start eating carbohydrate. Spontaneous changes in appetite for fat are always accompanied by inverse changes in the carbohydrate appetite. 4. The appetites for carbohydrate, fat, and protein, are dependetit to a large extent on the amount of the various components of the vitamin B complex present in the diet. By varying the combination of vitamin B components made accessible to the rats we are able almost at will to make a rat manifest a marked appetite or a marked aversion for either carbohydrate, fat, or pro- tein. In the absence of all vitamin B, rats eat large amounts of fat, little or no carbohydrate, and no protein. When thiamin chloride is made available as the sole representative of the vitamin B complex they eat carbohydrate, little fat, and still completely avoid protein. Progressively, as the other components are made available they eat more carbohydrate, less fat, more protein. When thiamin chloride, riboflavin, nicotinic acid, pyridoxin, pantothenic acid, choline chloride, and biotin are made available they thrive on a diet of very little fat, high carbohydrate, and moderately high protein. 5. Bulk apparently is not necessary. On the self-selection diets in which only purified substances were used, bulk was reduced to a minimum. The animals passed only a few small feces. The good health of the rats showed that bulk was unnecessary. Curt P. Richter 503 6. Diets self-selected from purified or nearly purified substances are more efficient than regular synthetic stock diets or diets composed of natural foods. Rats which selected their diets from sucrose, olive oil, casein, five mineral solutions, cod liver oil, and yeast ate approximately 35 per cent less food per day as measured in grams, than rats of the same age and weight on the stock diet. However, owing to the higher fat content of the self-selection diet, the caloric intake on the latter was only 18.7 per cent less than that of the animals on the stock food (Richter, Holt and Barelare"). Experiments on dietary selec- tions of pregnant and lactating rats offered the same choice of substances gave the most striking evidence for the efficiency of the self-selected diets. They showed a greatly increased appetite for protein (casein), fat (olive oil), calcium and phosphorus. At the height of lactation, rats on the self-selection diet nurs- ing the same number of young ingested only about half as much food as meas- ured in grams as rats on the stock diet. Despite the much lower food intake, the mother and young remained in excellent shape and the young were normal in size. Presumably the lowered food intake of the rats on self-selection diets is directly due to the ability to obtain the needed calcium, phosphorus, etc., with- out being forced to take increased amounts of the other unneeded components of the mixed diet (Richter and Barelare"). 7. Rats deprived of endocrine secretions luhich regulate metabolism of car- bohydrates, fats, proteins, or electrolytes, will when given the opportunity, select substances which compensate for the loss of the secretions. Thus, adre- nalectomized rats compensate for the increased excretion of sodium by in- gesting large amounts of sodium solutions and as a result keep themselves alive and free from symptoms of insufficiency. Likewise, parathyroidectomized rats manifest a greatly increased appetite for solutions of calcium, or for the closely related metals, magnesium and strontium (Richter and Eckert"). By virtue of the increased intake of these metals they not only survive but remain free from symptoms of tetany. Rats made diabetic by removal of the pancreas select substances, chiefly fat, which help to alleviate or even eliminate diabetic symp- toms. They actually gain weight, lose their polydipsia, polyphagia, and hyper- glycemia (Richter and Schmidt^). 8. The self-selection activities of rats may be used to determine functions of various glands and may serve to short-circuit prolonged biochemical investiga- tion. For example, it took biochemists many decades to determine that the adrenal glands play an important part in the regulation of sodium metabolism. Endless studies of the mineral content of the blood, urine, and feces were made before this result was achieved. With the self-selection technique, by offering adrenalectomized rats access to a variety of substances in purified form, we could have determined within a few days, and long before the advent of mod- ern biochemical methods, that the adrenal glands are concerned with the regulation of sodium metabolism. In the same way we could have determined that the parathyroids are concerned with the regulation of calcium metabo- lism. Thus far this method has indicated that riboflavin is concerned with the K04 Self-Selection of Diets metabolism of fat. Mannering, Lipton and Elvehjem'' have recently arrived at this conclusion on the basis of their biochemical studies. Our results have also indicated that protein metabolism is largely dependent on the vitamin B complex. 9. The self-selection technique may be used to bio-assay hormone and vita- min preparations. Thus, the increased calcium lactate appetite of parathy- roidectomized rats has been used to bio-assay vitamins which affect calcium metabolism. It was found that the calcium lactate intake could be reduced to approximately its normal level by 3,730 irradiated vitamin D units per kilo- gram body weight of vitamin D^ (crystalline); 8,880 units of irradiated ergos- terol; 3,380 units of vitamin D3 (crystalline); or by 3,770 units of irradiated cholesterol. Daily doses of dihydrotachysterol (A.T. 10) in minute amounts, 35 jLig. suffice to decrease the daily calcium lactate intake to its normal level. Parathyroid extract injected daily did not reduce the calcium lactate intake to its normal level until doses of 100 units per day were administered. This dosage was definitely toxic (Richter and Birmingham"). Similar assays have been made on the sodium regulatory functions of desoxycorticosterone using the sodium chloride appetite of adrenalectomized rats as a measure. 10. The maintenance of the total caloric intake at its normal level prevents the appearance of deficiency and other disturbances. It was found that when rats are forced to consume added calories with their drinking water they will reduce the caloric intake from their food in direct proportion to that taken from the other source. Thus, rats given a 16 per cent solution of alcohol as the sole source of water reduced their food intake directly in proportion to the caloric value of the ingested alcohol. This meant that the total caloric intake remained the same. That this actually was a beneficial selection is shown by the fact that rats which have been forced for over one hundred days to take 50 per cent of their total caloric intake in the form of alcohol did not develop any deficiency symptoms. It may be that in human beings similar effects from the use of alcohol depend on the fact that the caloric intake from the food added to that from the alcohol gives a total caloric intake which far surpasses the normal caloric requirements. If forced to take sugar in the same way, rats will reduce their food intake so that the total does not surpass the normal level (Richter'). 11. Solutions preferred in ayiy concentration to ivater by the rats are apt to have nutritional value, while those which are not preferred to water in any concentration are apt to have no nutritional value or to be toxic. Thus, rats were found to prefer solutions of the following substances to water: dextrose, maltose, levulose, sucrose, galactose, sodium chloride, potassium chloride, cal- cium lactate, alcohol— all of which are known to have a nutritional value. On the other hand, they were found to prefer water to solutions of the following substances: mercuric chloride, morphine sulphate, arsenic trioxide, phenyl- thiocarbamide— all of which are known to be poisonous (Richter and Camp- bell). Curt E Richter ^05 12. Taste thresholds may vary with internal Jieeds. Taste-threshold tests showed that normal rats were able to distinguish sodium chloride solution from water in concentrations of 1:2,000, while adrenalectomized rats, which have much lower blood-sodium values, due to the increased excretion of so- dium from the body, made the distinction in solutions of 1:33,000 (Richter and McLean^"). 13. Rats and Jniman beings have almost identically the same taste thresholds. Thus, the taste thresholds were practically the same for sodium chloride (0.055 per cent for rats, 0.087 P^^' cent for human beings); for sucrose (0.5 and 0.41 per cent respectively); for phenylthiocarbamide 0.0003 '^^^'^ 0.0003 P^^ cent respectively) (Richter and Clisby""). 14. The ability to make beyieficial selections apparently deperids npoyi taste, not on experience or the effects produced by the substances. After section of taste nerves, adrenalectomized rats no longer selected salt. The lower taste threshold of adrenalectomized rats also brings evidence to support this view. Rats showed an increased appetite for salt solution in concentrations which could not possibly have had any physiological effect. \^. Inability to ?nake beneficial selections may depend oji inherited taste dejects, that is, 07t the inability to taste certain substances. Salmon and Blakes- lee^ and Blakeslee"* showed that a small percentage of people are unable to taste the bitter phenylthiocarbamide. We have found many individuals who are unable to taste sucrose in concentrations as high as 16 percent, or riboflavin in crystalline form. Blakeslee and Fox"" and Snyder^ reported that inability to taste phenylthiocarbamide is inherited as a Mendelian recessive. Some of our observations indicate that the inability to taste riboflavin and alcohol may also be inherited. 16. All of the various types of self-selection behavior form part of Claude Bernard's concept of the constancy of the internal environment. Bernard^^ and Cannon^" pointed out and enumerated physiological mechanisms which help to maintain homeostasis, or a steady state. These self-selection experiments have shown that behavior mechanisms, that is, responses of the total organism, are used to maintain a constant internal en\ ironment (Richter^''). REFERENCES 1. Evvard. J. M.: Proc. Iowa Acad. Scis. 22:375, 1915- 2. Evvard, J. M.: Iowa Agricult. Exper. Statn. Bull. no. 1 18, 1929. 3. Pearl, R., and Fairchild, T. E.: Amer. Jl.Hyg. 1:253. i92i- 4. Dove, W. P.: Amer. Natiirlst. 69:469, 1935. 5. Le^vis, S. L.: .\mer. Pigeon Jl. 16:21 1, 1927. 6. Stearns, G. L., and Hollander, W. F.: Amer. Pigeon Jl. 28:355, 1939. 7. Godden, W.: Jl. Agric. .Sci. 16:78, 1926. 8. Orr, J. B.: Minerals in Pastures (London: 1929). 9. Theiler, A.; Green, H. H., and \'iljoeii, P. R.: Repts. Director \'eterin. Resrch., I'nion South Africa 3/4:9, 1915- 10. Oshorn, T. B., and Mendel, L. B.: Jl. Biol. Chem. 20:351, 1915. 11. Mitchell, H. S., and Mendel, L. B.: Amer. Jl. Physiol. 58:211, 1921. 5o6 Self-Selection of Diets 12. Davis, C. M.: Amer. Jl. Diseases Childr. 36:651, 1928. 13. Davis, C. M.: Amer. Jl. Diseases Childr. 46:743, 1933. 14. Sweet, C: Jl. Amer. Med. Assn. 107:765, 1936. 15. Richter, C. P., and Havvkes, C. D.: Amer. Jl. Physiol. 131:639, 1941. 16. Richter, C. P.; Holt, L. E., Jr.; Barelare, B., Jr., and Hawkes, C. D.: Amer. Jl. Physiol. 124:596, 1938. 17. Richter, C. P.; Holt, L. E., Jr., and Barelare, B., Jr.: Amer. Jl. Physiol. 122:734, 1938. 18. Richter, C. P., and Barelare, B., Jr.: Endocrinology 23: 15, 1938. 19. Richter, C. P., and Eckert, J. F.: Amer. Jl. Med. Scis. 198:9, 1939. 20. Richter, C. P., and Schmidt, E. C. H., Jr.: Endocrinology 28:179, 1941. 21. Mannering, G. J.; Lipton, M. A., and Elvehjem, C. A.: Proc. Soc. Exper. Biol. & Med. 46: 100, 1941. 22. Richter, C. P., and Birmingham, J. R.: Endocrinology 29:655, 1941. 23. Richter, C. P.: Quart. Jl. Studies Alcoh. 1 :650, 1941. 24. Richter, C. P., and Campbell, K. H.: Jl. Nutrit. 20:31, 1940. 25. Richter, C. P., and MacLean, A.: Amer. Jl. Physiol. 126: 1, 1939. 26. Richter, C. P., and Clisby, K. H.: Amer. Jl. Physiol. 134: 157, 1941. 27. Salmon, T. N., and Blakeslee, A. F.: Proc. Natn. Acad. Scis. 21:78, 1935. 28. Blakeslee, A. F.: Proc. Natn. Acad. Scis. 18:120, 1932. 29. Blakeslee, A. F., and Fox, A. L.: Jl. Hered. 23:97, 1932. 30. Snyder, L. H.: Ohio Jl. Sci. 32:436, 1932. 31. Bernard, C: Legons sur les Proprietes Physiologiqiies et les Alterations Pathologiques des Liquides de I'Organisme (Paris: 1859). 32. Cannon, W. B.: The Wisdom of the Body (New York: 1932). 33. Richter, C. P.: Psychosom. Med. 3:105, 1941. THE RELATIONSHIP OF THE ANTERIOR PITUITARY TO THE THYROID AND THE ADRENAL CORTEX IN THE CONTROL OF CARBOHYDRATE METABOLISM By JANE A. RUSSELL X From the DEPARTMENT OF PHYSIOLOGICAL CHEMISTRY YALE UNIVERSITY SCHOOL OF MEDICINE NEW HAVEN, CONNECTICUT THE RELATIONSHIP OF THE ANTERIOR PITUITARY TO THE THYROID AND THE ADRENAL CORTEX IN THE CONTROL OF CARBOHYDRATE METABOLISM IN THE PAST several years, the anterior pituitary gland has been shown to have important functions in the control of the intermediary metabolism of carbo- hydrate and of protein, and directly or indirectly in that of fat. The determina- tion of the precise role of any hormone in metabolism is difficult, but in the case of the anterior pituitary it is the more so because of the manifold activities of this gland. Particularly, complications arise from the trophic control by the anterior pituitary of two other endocrine organs which are known to affect certain phases of metabolism— the thyroid and the adrenal cortex. Thus, the removal of the hypophysis brings about extensive atrophy of these glands, and extracts of the anterior pittutary contain factors which affect the activity of these glands and which are to be separated, chemically or physiologically, from those other factors which may exert direct effects on metabolic activities. Much of the work of recent years which has sought to define the nature of hormonal control of metabolism has of necessity been devoted to the inter- relationships between the hormones of the anterior pituitary, thyroid, and adrenal cortex. Since most of these studies have been in the field of carbo- hydrate metabolism, this discussion will be confined to that subject. It must be remembered, however, that whenever changes occur in the metabolism of carbohydrate inevitably changes also are to be expected in fat and often also in protein metabolism. These may be very important, or they may even be primary to alterations in carlDohydrate metabolism; it is only because of a scantiness of data bearing on the interrelations of the hormones in these fields that they are inentioned here but briefly, as they seem related directly to the topic in hand. The chief effects of the removal of the anterior pituitary gland or of treat- ment with anterior pituitary extracts on metabolism are now familiar phe- nomena and many phases of these subjects have been reviewed."""^ The otu- standing facts may be again outlined here, before they are analyzed with respect to the various hormonal factors affecting them. The first strong indica- tion of a relationship of the anterior pituitary to carbohydrate metabolism was the observation by Houssay, in 1924, of the remarkable hypersensitivity to instdin of hypophysectomized animals. This discovery was followed by that which is now called the Houssay effect— the amelioration of pancreatic dia- betes by hypophysectomy. Then the diabetogenic effects of anterior pituitary extracts were demonstrated— the induction of states simulating pancreatic dia- C 509 ] ^ 1 o Pituitary and Carbohydrate Metabolism betes by the administration of APE* to so-called "Houssay" animals and later to normal animals of many species. Permanent diabetes, lasting long after the cessation of APE injections, was first observed by Evans and collaborators and later extensively studied by Young,' Campbell and Best,'' Dohan, Fish and Lukens," and others. This permanent diabetes, observed so far only in dogs, has now been ascribed to damage to the pancreatic islets induced by unknown mechanisms during the prolonged administration of APE. The temporary diabetes first observed, however, continues to be considered of direct pituitary origin. All these observations have been many times confirmed, and many other metabolic effects of APE have also been demonstrated— the glycotrophic or anti-insulin effect (in which all trace of insulin action may be prevented by suitable preparatory administration of APE to the experimental animals), the pancreotrophic effect, the power of increasing the content of liver fat by trans- ference of fat from peripheral depots, the ketogenic factor, and the glycostatic effects— among them. In recent years, in work on small animals such as the rat and rabbit, it has been possible to investigate the role of the anterior pituitary in metab- olism with more precision than is usually possible in larger animals, such as the cat and dog, in which most of the earlier work was done; and most of the work defining the interrelationships of the hormones has been carried out on the smaller animals. In the rat, rather complete studies have been made of the effects of hypophysectomy upon the metabolism of carbohydrate. Hypophysec- tomized rats, like other species of such animals, exhibit atrophy of the thyroid, adrenals, and gonads, and fail to grow, but as long as they are well fed, they do not show any great abnormalities in their metabolism of carbohydrate except an extreme hypersensitivity to insulin. However, when these animals are fasted, differences from the normal are at once apparent. In as short a time as eight hours after the removal of food, glycogen is found to have disappeared almost entirely from the liver, and the blood sugar has begun to fall rapidly. At about this time the muscle glycogen content begins to fall, and by the time the animal has been fasting 16 to 18 hours, the muscle glycogen may be only half its normal value, whereas of course in normal animals the muscle glycogen is ordinarily quite stable for long periods of fasting.'' During this period of carbohydrate depletion, the RQ of the hypophysectomized rats is higher than that of the normal, and sufficiently so to indicate that the rapid rate of glycogen loss could have been due to more rapid oxidation of carbohydrate.' Similar studies have not as yet been completed in other species than the rat; but the loss of liver glycogen and rapid decline of the blood sugar has been observed frequently. Cope' and others have shown clearly in hypophysectomized rabbits that a rapid loss of liver glycogen and subsequent decline in the blood sugar sets in during fasting and something like this situation also is found in hy- * APE refers to anterior pituitary extracts, without distinguisliing between any of the various types of extract. In most cases a crude, or at least not highly purified, extract was used, which contained many of the several anterior pituitary factors. Jane A. Russell ^ 1 1 pophysectomized cats/" but in these cases the changes occur more slowly than in rats. In the fasted hypophysectomized rat the abnormal loss of muscle glycogen can be prevented entirely by APE administration and the RQ is depressed to normal at the same time.^ " '- This action— the maintenance of fasting muscle-glycogen levels in hypophysectomized animals— has been de- fined as the glycostatic effect, and it can be used to assay APE. The term is also sometimes applied to the effects of APE on liver- and blood-carbohydrate levels, in normal as well as in hypophysectomized animals, but, as will be evident later, these data do not always afford trustworthy evidence of glyco- static activity. Two possible explanations exist for the apparent rapid rate of disappear- ance of carbohydrate in the absence of the pituitary and its prevention by APE. One is, of course, that the rate of oxidation of carbohydrate is increased. The other is that, since ordinarily the carbohydrate levels are maintained during fasting by gluconeogencsis, interruption in this process occurs in hy- pophysectomized animals and a normal rate of peripheral oxidation permits the apparently more rapid disappearance of carbohydrate. However, there is not a sufficient fall in the rate of excretion of nitrogen after hypophysectomy in the rat to make a critical difference in gluconeogencsis probable. In addi- tion, data obtained from experiments on rats fed glucose and on eviscerated animals indicate that an accelerated rate of peripheral oxidation of carbo- hydrate exists in the hypophysectomized animal. One of the first, but still most convincing, of these experiments was one in which hypophysectomized and normal rats, after being fasted, were fed meals of carbohydrate (starch) and the glycogen and blood-sugar values followed for 36 hours thereafter. It was found that the hypophysectomized rats could be fed twice as much carbohydrate as the normal rats and at the end of the experimental periods still have much lower levels of liver and muscle glycogen than the control animals. That is, despite metabolic rates one-third below normal, the hypophysectomized rats must have disposed of twice as much carbohydrate as the normal animals in the given time.' Further demonstration of the increased rate at which hypophysectomized animals dispose of carbo- hydrate is seen, together with an indication of its fate, in balance studies of the Cori type (table i).'" In this case, a much smaller proportion of the absorbed glucose was reco\ered 4 hours after its feeding in the hypophysec- tomized rat than in the normal, and respiratory data obtained during the absorption period indicated that the carbohydrate which disappeared could have been oxidized. Now if one is to explain the difference between the find- ings in the normal and the hypophysectomized rats on the basis only of a decrease in gluconeogencsis in the latter case, then one would have to assume that massive gluconeogencsis had taken place in the normal animals under these conditions— in an amount equal to about half of the carbohydrate de- posited—in the presence of a plethora of carbohydrate. 512 Pituitary and Carbohydrate Metabolism Anterior pitviitary extract, administered to normal or to hypophysectomized rats, in experiments of exactly similar type to those described above, reverses the direction of the changes. In the hypophysectomized animal, the restora- tion to normal by APE is not quite complete, but in normal animals the changes are striking. There is a marked fall in the RQ and increased deposi- tion of muscle glycogen, but not of liver glycogen. This fall in the RQ in glucose-fed animals has been used as the basis of assay of APE for metabolic effects."" TABLE 1 Disposition of Fed Glucose in Rats (4 hours after feeding) No. of obser- vations Carbohydrate recovered (per cent of absorbed glucose) as Carbohydrate oxidized (calculated from l^issiie glucose Muscle glycogen Liver glycogen Total respiratory data) per cent of absorbed glucose Normal rats .... 19 7 3 2 5 l6±l* 9 + 2 27±4 i6.5±o.7 6.5±o.7 35-5 17-5 -18 49.2 + 14 49-5±i-3 68. j; 4-2.0 + 19 " 28.8±2.4 — 21 Hypophysectomized rats. Difference from normal Normal rats given APE f. . Difference from normal 17. 2 £2.4 * Standard error. t Saline anterior pituitary extract, I ml. given intraperitoneally I to 2 hours before glucose feeding. Additional evidence that the anterior pituitary affects peripheral oxidation of carbohydrate has been obtained in eviscerated or hepatectomized animals. The first positive indication that APE had a peripheral action was obtained by Marks,^^ who found that it prevented the deposition of muscle glycogen following insulin administration in eviscerated cats. Himsworth and Scott^'^ continued this work by determining the relative effects of APE and insulin on the survival time of functionally hepatectomized rabbits. The data in table 2, taken from their paper, where the time between hepatectomy and the occurrence of hypoglycemic convulsions measures inversely the rate of fall of the blood sugar, shows that the APE prevented the hypoglycemic effects of insulin in the absence of the liver. More recently, the metabolism of the hypophysectomized eviscerated prepa- ration has been studied. Greeley" first determined that hypophysectomized rabbits, in which the blood sugar falls very rapidly when they are fasted, re- quire the administration of glucose at the rate of about 500 mg. per kilogiam per hour to maintain normal blood-sugar levels. Of course, the normal rabbit requires no extra glucose beyond that which it makes itself. Now this large glucose requirement has been found to persist unchanged in the absence of the liver or of the abdominal viscera, and it is several times the amount needed by the normal eviscerated rabbit (Drury'*). This abnormally great Jane A. Russell 513 peripheral use of glucose after hypophysectomy has been found now also in the eviscerated rat, and in addition it has been shown not to be due to in- creased deposition of muscle glycogen or to lactic acid formation (Russell"). In fact, in the absence of added glucose, a loss of muscle glycogen occurs in TABLE 2 Anterior Pituitary Extract and Insulin in Hepatectomized Rabbits (From Himsworth and Scott) With insulin only With APEt only With APE and insulin No. of observations Survival time,* minutes 82 ± 5 I79±i4 i8o±i6 * Time after hepatectomy until convulsions occurred, t Young's giycotrophic extract. the hypophysectomized eviscerated rat (but not in the normal animal), which can be prevented by pretreatment of the animal with APE, just as in the intact fasted rat. The rate of fall of the blood sugar after evisceration in the rat is also increased by hypophysectomy and diminished by APE (table 3). TABLE 3 Carbohydrate Metabolism in Eviscerated Rats Normal rats (adrenal-demedullated) . Hypophysectomized rats Hypophysectomized rats given APE Adrenalectomized rats (salt treated) Rate of fall in blood sugar, mg. per cent per hour 37 i 3 • o 72^5.8 47±3-6 58±6.7 Rate of change in muscl e glycogen. mg. P per cent ;r hour + 2ill — I o8±i6 + I ±13 — I± 7 Glucose required to maintain normal blood sugar level, mg. per lOO gm. per hour I3.5-I4.O 26 — "SO Thus, since the rapid rate of disappearance of carbohydrate noted after hy- pophysectomy persists in the absence of the liver, which is the prime site of gluconeogenesis, it can be assumed that there exists an accelerated rate of peripheral oxidation of carbohydrate in the absence of the pituitary. Many other changes in metabolism observed after hypophysectomy can be associated with a high rate of oxidation of carbohydrate in conditions where ordinarily carbohydrate oxidation is reduced to a minimum— among them the failure to recover from insulin hypoglycemia, the fatal hypoglycemia and relative failure of fat and protein metabolism which is found in phlorizinized hypophysectomized animals, and the apparent resumption of carbohydrate metabolism in the absence of the pancreas. Yet, in none of the latter cases is 514 Pituitary and Carbohydrate Metabolism it possible to say whether increased peripheral oxidation is the sole explana- tion of these changes. Evidence is available that a diminution of gluconeo- genesis occurs in the diabetic states after hypophysectomy, and increased gluconeogenesis is associated with the adrenotrophic and with the anti-insulin activities of certain APE. Thus it is possible that the anterior pituitary pro- TABLE 4 Carbohydrate Levels in Thyroidectomized-Parathyroidectomized Rats No. of observa- tions Glucose absorp- tion, mg. per 100 gm. per hour Blood glucose, mg. per cent Liver glycogen, per cent Muscle glycogen, mg. per cent Oxygen consump- tion, ml. per 100 gm. per hour RQ (uncorrected) A. Not fasted Normal rats •3 5 125 92 1-7 3-7 53° 5V Thyroidectomized rats B. Fasted 24 hours Normal rats Thyroidectomized rats 15- 12-19 72 82 0. 1 0. 1 524±IO 465±i2 I54±2 IOO±2 .725zt .005 .7i9±.oo3 C. Fed glucose after 24- hour fast (4 hours after feed- ing) Normal rats Thyroidectomized rats Thyroidectomized rats given APE. . 19 8 9 185 96 89 127 120 122 2.98 1.26 1.27 748±i8 584±io 667±i2 I45±2 97±3 1 1 1±5 .887-f-.oo5 . 826±.ooi; .778±.oo5 duces some part also of its effects on carbohydrate metabolism by affecting gluconeogenesis. Since the adrenal cortical hormone is now known to pro- mote gluconeogenesis and since thyroid activity may also limit the rates of this and other intermediary metabolic reactions, the roles of the thyroid and adrenal hormones in metabolism must next be examined. The Thyrotrophic Relaiionsliip: One of the first trophic effects of the an- terior pituitary to be established was that on the thyroid. In view of the well- known effects of the thyroid on the metabolic rate of tissues, and therefore presumably upon the rates of many metabolic reactions, it would be reason- able to suppose that some of the effects of the anterior pituitary might be mediated through this gland. However, few comprehensive studies seem to have been made on the metabolism of thyroidectomized animals or on ani- mals made only moderately hyperthyroid with thyroid hormone, so that it is Jane A. Russell 515 difficult to compare in a general manner the effects of the thyroid to those of the anterior pituitary on carbohydrate metabolism. Of the experimental observations which have been made in this field, most fall into the following classes: (a) loss of glycogen (and fat) reserves in animals given large amounts of thyroxin, due in all probability to rapid utilization in the hypermetabolic state; (b) abnormalities in glucose tolerance cinves, the "diabetic" type being T.^BLE 5 Dispo.siTiON OF Fed Glucose in Thyroidectomized Rat.s Four Hour.s after Feeding (Calculated from table 4) Percentage of absorbed glucose Found as Tissue glucose Muscle glycogen Liver glycogen Total recovered ().\idized I. Normal rats 3 5 6 16 16 29 16.5 10.2 12.6 35-5 31.2 - 47-6 - 4-3 + 16.4 49-5 sO.O ,U.8 + 0. 9 + 1.*; 0. > s'th 6 7 ^ 14 +6.5 0. s ;:h ^ 6 0 6 +0.3 0.5 ^th 5 6 - +0-7 Average +3-1 anterior-pituitarv extracts ^\•hich caused luteinization of the ovaries. Pincus," also reported a prolongation of pregnancy in lactating mice by injections of estrone-free corpus Itueum hormone during the five to eight days after the first parturition. He concluded that the excessive corpus luteum secretion caused the delayed implantation during lactation. The source of corpus luteum hormone used in the present work -vvas "Lu- tren." a preparation of synthetic progesterone in oil, manufactured by Bayer I. G. Farbenindustrie (Leverkusen, Germany). The results are represented in table 1. From this table it appears that when progesterone "^vas given during lacta- tion, before the expected implantation, from the third to the seventh day after the first delivery, a prolongation of pregnancy ensued, which in all cases except one (mouse no. 26) was longer than anticipated if only lactation had caused the delav. TJie Effect of Hypophysial Lactogefiic Hormone upon Prolongation of Pregnancy During Lactation. Originally, anterior pituitary lactogenic hor- mone was connected with the process of lactation. Later several new properties were ascribed to this same hormone. DreseP" observed a distinct suppression 568 Prolongation of Pregnancy of estrous cycles in mice when lactogenic hormone was injected. Lahr and Riddle" were able to demonstrate the same thing in rats. They inclined to the view that the amenorrhea in mammals and human beings during pregnancy and lactation is due to the presence of lactogenic hormone. Nathanson, Fevold and Jennison"'^* believed that the experimental suppression of the estrous cycle in mammals is due to the presence of luteinizing hormone in the com- mercial preparations of lactogenic hormone. Evans, Simpson, Lyons and TABLE 2 Effect of Lactogenic Hormone on Prolongation of Pregnancy During Lactation Mouse No. 74- 95- 21 . 134- 64. 26. 135- 121 . 35-^ lb 31^ Dose iniected, B.U. 5 5 5 5 5 5 ID ID ID* ID* 20* Day of injection 3d 3d 4th 4th 6th 7th 4th 6th 2d-8th 2d-8th 2d-8th Number of suckling young 5 7 4 8 6 4 3 6 <; 6 5 Prolongation of pregnancy Expected, days Observed, days 7 1 1 1 1 no litter 17 8 no litter no litter 10 no litter no litter Difference, days + 0.7 + 2.3 + 5-9 +9-5 + 2.9 +3-7 Average +4- 2 * Divided in two daily doses. Turpeinen^'^ were able to produce deciduomata in hypophysectomized rats with injections of pituitary lactogenic hormone. They were of the opinion that lactogenic hormone increases the production of corpus luteum hor- mone thus sensitizing the uterine mucosa for the deciduoma reaction. This and the existence of functional corpora lutea during lactation, as well as the release of hypophysial lactogenic hormone at delivery, made it worth while to study the question whether lactogenic hormone injections during lactation had the same effect on the delay of implantation as the corpus lu- teum hormone. In these experiments lactogenic hormone of the anterior hypophysis was used in the form of "Prolactin" manufactured by Eli Lilly %:. Company (In- dianapolis, Ind.). This preparation is a yellowish powder, which is standard- ized in pigeons (B.U.) according to the method described by Bates, Riddle and Lahr.""" For these experiments it was dissolved in distilled water. The re- sults appear in table 2. The results show conclusively that injections of lactogenic hormone into Kaisa Turpeinen 569 lactating and pregnant mice prolong the span of gestation still more than lactation alone, and given in large doses may entirely prevent pregnancy, presumably because of too excessive delay of implantation. Discussion There seems to be a reciprocal effect in the relationship between lactation and the implantation process. From the work of Frankl"^ we know that trans- plantation of placentas into pregnant animals near term inhibits lactation. On the other hand, lactation has an inhibiting effect on implantation by delaying it. The action of lactogenic hormone and corpus luteum hormone prevails during lactation. These facts and the results of my work support the view that the excessive production of lactogenic hormone and corpus luteum hormone during lactation wotdd be the cause of delayed implantation. Summary The prolongation of pregnancy, which occurs in some species of lactating ro- dents and which is due to the delay of implantation, has been studied. Particu- lar attention has been given to the effect on the prolongation of pregnancy of progesterone and hypophysial lactogenic hormone administered during the first part of the pregnancy, that is during the implantation time. Albino mice were used as experimental animals. The hormones were injected subcuta- neously, and their effect on the prolongation was noted. Before starting the experiments with hormones, the effect of the number of young on prolongation of pregnancy during lactation was studied. The following conclusions were reached: 1. In accordance with the majority of earlier workers, it was found that the length of pregnancy in lactating mice is dependent on the number of young the mother is suckling. There is an almost perfect positive correlation between these two variables (r= -1-0.928). Each of the young lactated prolongs the pregnancy on the average by 1.2 days. 2. There is no demonstrable correlation between the delay and the number of young carried. 3. The synthetic corpus luteum hormone (progesterone) causes a further prolongation of pregnancy, that is, the length of pregnancy is prolonged be- yond what lactation alone would cause. 4. The injection of the hypophysial lactogenic hormone causes further pro- longation of gestation during lactation. Given in larger doses it may entirely prevent littering. I wish to express my gratitude to Dr. Carl G. Hartman for suggesting this problem and for his aid during the early part of the work. Since Dr. Turpeinen's return to her fatherland, it has been found by two experimenters. Dr. Segar-Jones at the Johns Hopkins Medical School, working with the mouse and rat, and 570 Prolongation of Pregnancy by Dr. C. K. Weichert working Avith the rat (for the latter see Anat. Recrd. 8i[supp.]:28, 1941), that estrogens favor the attachment of the vesicle at normal time and shorten the gestation period of the lactating female to normal. REFERENCES i.Lataste, F.: Cpts. rd. Soc. de biol., mems. 43:21, 1891. 2. Brambell, F. W. R.: Amer. Jl. Obstet. & Gynecol. 33:942, 1937. 3. Daniel, F. J.: Jl. Exper. Zool. 9:865, 1910. 4. King, H. D.: Biol. Bull. 24:377, 1913. 5. Kirkham, W. B.: Anat. Recrd. 11:31,1916. 6. Kirkham, W. B.: Jl. Exper. Zool. 27:49, 1918. 7. Long, J. A., and Evans, H. M.: Anat. Recrd. 18:241, 1920. 8. Mirskaia, L., and Crew, F. A. E.: Proc. Roy. Soc. Edinburgh 51:1, 1930-31. 9. Enzmann, E. N.; Saphir, N. R., and Pincus, C: Anat. Recrd. 54:325, 1932. 10. Hain, A. M.: Quart. Jl. Exper. Physiol. 24:101, 1934. 1 1. Teel, H. M.: Amer. Jl. Physiol. 79:170, 1926. 12. Wislocki, G. B., and Goodman, L.: Anat. Recrd. 59:375, 1934- 13. Hamlett, G. VV. D.: Quart. Rev. Biol. 10:432, 1935. 14. Pincus, G.: The Eggs of Mammals (New York: 1936). 15. Fisher, R. A.: Statistical Methods for Research Workers (7th ed.; Edinburgh, London: 1938). 16. Dresel, L: Science (n.s.) 82:173, 1935. i7.Lahr, E. L., and Riddle, O.: Proc. Soc. Exper. Biol. & Med. 34:880, 1936. 18. Nathanson, I. T.; Fevold, H. L., and Jennison, D. B.: Proc. Soc. Exper. Biol. & Med. 36: 481, 1937- 19. Nathanson, I. T., and Fevold, H. L.: Endocrinology 22:86, 1938. 20. Evans, H. M.; Simpson, M. E., and Turpeinen, K.: Anat. Recrd. 70 (supp. 3):26, 1938. 21. Evans, H. M.; Simpson, M. E.; Lyons, W. R., and Turpeinen, K.: Endocrinology 28:933, 1941. 22. Bates, R. W.; Riddle, O., and Lahr, E. L.: Amer. Naturlst. 69:55, 1935. 23. Frankl, O.: Amer. Jl. Obstet. & Gynecol. 6:399, 1923. IS INCREASED CAPILLARY FRAGILITY A SIGN OF ASCORBIC ACID SUBNUTRITION? By OSMO TURPEINEN From the INSTITUTE OF PHYSIOLOGY, UNIVERSITY OF HELSINKI HELSINKI, FINLAND IS INCREASED CAPILLARY FRAGILITY A SIGN OF ASCORBIC ACID SUBNUTRITION? IT IS WELL KNOWN that the tendency to bleed is one of the main symptoms of scurvy. In manifest scurvy the hemorrhages either occur spontaneously or are caused by trivial traumas. This symptom is not based on any pathologic changes in the composition of the blood, for both the bleeding time and the coagulation time are normal, but on decreased capillary resistance, which apparently is due to defective formation of intercellular substances. This symptom of increased capillary fragility, which usually is the first morbid expression of scurvy, has also been utilized for diagnosis of milder deficiency states, so-called subclinical scurvy. In this condition the hemor- rhagic tendency is not severe enough to lead to manifest symptoms. It can, however, be demonstrated by subjecting the capillaries to extra strain. This may be done by using the well-known Rumpel-Leede test, which in manifest scurvy is, as a rule, strongly positive.* In an attempt to refine the Rumpel-Leede test and to bring it to a more quantitative basis, Gothlin^ some years ago developed a capillary-fragility test which he thought could be utilized for diagnosis of subclinical scurvy. This method consists of increasing the intracapillary pressure in the arm by means of a Riva-Rocci cuff and counting the petechial hemorrhages which appear in a specified skin area under standard conditions. The original procedure has since been modified." Besides this "overpressure" method, another form of the capillary-fragility test has also been in use. This is the "suction" or "underpressure" test, originated by Hecht' and later used and modified by several workers (see Jersild and Elmby^). The two methods do not necessarily give concordant results. According to Gothlin" the overpressure method is preferable for its basis is "clear and unassailable," which "can by no means be said of the suction procedure." After the chief role of ascorbic acid in the etiology of scurvy had been demon- strated, it was generally assumed that the increased capillary fragility as shown by the above-mentioned tests would, in otherwise healthy persons, be indica- tive of ascorbic acid subnutrition. Accordingly, these methods have gained a wide use in assessing the nutritional status in regard to ascorbic acid. It is not, however, quite clear how reliable a sign the increased capillary fragility is in showing ascorbic acid deficiency. There is evidence both pro and con. Gothlin and his collaborators, who are the chief proponents of the test, have furnished most of the evidence in favor of the test.^''''"'^ Some con- firmative evidence has also come from other quarters; thus Dalldorf," Molitch,^ and Cianci'" considered the test satisfactory. Also Sloan" found the capillary- * However, cases of scur\y have been repored which, even in spite of the presence of spontaneous hemorrhages, give negative results in capillary-fragiUty tests.^"* [573] 574 Ascorbic Acid Subnutrition fragility test useful though less sensitive than chemical methods in indicating ascorbic acid subnutrition. On the other hand, there are many dissenting opinions. Several investiga- tors"'"'' have for various reasons questioned the reliability of the test. Weld* found that vitamin D was even more effective than ascorbic acid in increasing capillary resistance. Thus the opinions about the capillary-fragility test are rather contradictory. In view of the frequent use of this test in nutritional surveys, it would, how- ever, seem desirable to have at hand some more definite evidence in regard to its reliability as a means of detecting ascorbic acid subnutrition. The pres- ent paper is a further attempt to evaluate the capillary-fragility test. Correlation of Capillary Fragility ivith the Ascorbic Acid Content of the Diet. If ascorbic acid deficiency were a cause of increased capillary fragility, a correlation should exist between the ascorbic acid content of the diet and the capillary resistance. This question has not to my knowledge been studied previously on any considerable scale. Gedda'° and Nordenmark," it is true, demonstrated the existence of a seasonal difference in capillary fragility, which apparently paralleled a corresponding difference in ascorbic acid intake. The same fact is also evident from my data, which indicate a very definite seasonal difference between the capillary resistance values in the spring and those in the late autumn and an equally definite, parallel difference in the amounts of dietary ascorbic acid. Such evidence, however, is of limited value, for it is clear that a multitude of dietary and other environmental factors undergo seasonal variations. Thus there is not enough justification for the conclusion that the seasonal changes observed in capillary resistance were due to changes in ascorbic acid intake. On the other hand, it would constitute decidedly more valuable evidence in favor of the test, if it were possible to demonstrate a fair correlation between these two variables in a sufficiently large body of material. An extensive mass of data collected by the Finnish National Nutri- tion Committee some time ago offered an opportunity for studying this question. In the years 1936-1937 the diet of a number of families of the lower income classes in Finland was examined. All food consumed by the families during a lo-day period was carefully weighed, and its contents of the chief nutrients, ascorbic acid among them, were calculated. The amounts of the nutrients found in the diet were compared with the estimated requirements of the family. The requirements for ascorbic acid were taken as 50 mg. per day for persons over ten years of age and 25 mg. per day for children less than ten years old. Capillary-fragility tests were carried out by Gothlin's method® on the members of the families (616 persons in all), the results being expressed as the number of petechiae developing under 50 mm. (Hg) pressure. Table 1 shows the distribution of the 616 persons examined in regard to their ascorbic acid intake and the capillary status. This table reveals that the Osmo Turpeinen 575 correlation between these two variables cannot be a very close one, for the values seem to be rather irregularly distributed. Among families with very low intakes of ascorbic acid, cases of normal capillary resistance are often found, and, on the other hand, a relatively high ascorbic acid intake does TABLE 1 Distribution of 6i6 Persons in Regard to the Ascorbic Acid Content of the Diet and Capillary Fragility Dietary ascorbic acid — percent of the estimated requirement Totals Array means. Number of petechiae* 20- 40 40- 60 60- So 80- 100 100- 120 120- 140 140- 160 160- 180 180- 200 200- 220 220- 240 240- 260 per cent ascor- bic acid O I 2 ,) 4 5 6 7 8 9 lO II 12 13 Over 13 3 2 I 5 I 3 2 3 10 8 7 4 3 21 13 15 7 8 6 4 6 5 6 I 7 18 0 8 4 2 I 3 32 16 14 9 8 6 2 3 2 I I I 2 3 50 20 23 17 6 5 3 4 3 3 I 3 16 9 2 I I I I 2C 5 ) ^ 6 7 5 5 2 2 I I 3 1- 12 10 4 5 4 2 I I c I 206 109 86 56 41 28 16 15 14 10 8 3 2 6 16 127 131 119 112 129 129 "3 103 97 98 142 97 80 63 92 Totals . 20 36 99 56 100 141 31 48 29 12 41 3 616 Array means, pete- chiae . . 4-9 2-3 5-2 2.9 3-0 2.6 1.4 1-7 2.2 0.8 2-3 0.3 * The number of petechiae indicates: 0-4, normal capillary resistance; 5-8, borderline cases; and over 8, increased fragility (Gothlin). not seem to exclude increased capillary fragility. Only when the ascorbic acid intake reaches the level of nearly twice the estimated requirement do cases of increased capillary fragility disappear. However, if the array means shown in table i are examined, it may be seen that, though there is much irregular fluctuation, the capillary-fragility values tend to decrease slightly when ascorbic acid intake is increased, and, on the other hand, the average intake of ascorbic acid generally becomes somewhat lower in persons with higher capillary fragility. Thus there seems to be at least some kind of cor- relation between these two variables. 57^ Ascorbic Acid Subniitrition To define the degree of this correlation more accurately, a correlation co- efficient was computed from the data given in table i. This was found to be: r = — 0.155 =t 0.040 (S.E.). As this coefficient is nearly four times its standard error, it is certainly statis- tically significant. Therefore, a negative correlation undoubtedly exists be- tween the ascorbic acid intake and the number of petechiae in Gothlin's test. This correlation is, however, of fairly low degree. If compared with some known correlations (Pearl"*), it is, for example, seen to be no better than the correlation between the brain weight and the body weight and only slightly better than that between intelligence and the horizontal head circumference. There are, however, factors which may have unduly lowered this coefficient. First, the regression of the number of petechiae on the ascorbic acid content of the diet need not necessarily be rectilinear; some other function might give a better fit and, conseqtiently, a higher correlation.* In view of the irregular course of the array means (table 1), it seems very doubtful, however, whether any simple function could be found which would give an essentially better fit with the data. In any case, the maximum influence of this factor may be estimated by computing the correlation ratio, a value which provides an upper limit such that no regression function giving a higher correlation can be found. This was found to be: rj = 0.225. This is higher than the correlation coefficient but still a relatively low value. Furthermore, the ascorbic acid values may be somewhat inaccurate because individual differences may have existed in the ascorbic acid intake among the different members of the families, differ- ences which have been disregarded by using the familial averages of the as- corbic acid intake values for the individuals. These two factors may have exerted a somewhat depressing action on the value of the correlation co- efficient. Their influence, however, can hardly have been great enough to alter significantly the fact that the correlation between the ascorbic acid content of the diet and the capillary resistance is low. Effect of Treatment with Ascorbic Acid on Capillary Fragility. The signifi- cance of ascorbic acid as a capillary controlling factor should also manifest itself in the capacity to raise an abnormally low capillary resistance, and to prevent a normal resistance from falling. The earlier experiments on this point have yielded discordant i-esults.*'^"'"'"'^®'^""'^ Besides, many of these experi- ments have been inadequately controlled and on a relatively small scale. The problem was further studied on the basis of the data next presented.-j" In March, 1941, the capillary resistance of a number of elementary school children was measured by Gothlin's improved method," the double-sided petechial index being determined. From these children two equal groups * See Ahlborg and Brante," who found that the relationship between the phisma ascorbic acid and the capillary fragility expressed as the "petechial index" could best be represented by a hyperbola. f This part of the work was aided by a grant from the foundation "Suomen Kulttuurira- hasto." Osmo Turpeinen 577 were formed. One of these received loo mg. of synthetic ascorbic acid daily for fifty days, the other group being held as the control. During this period the children lived at their homes and continued on their usual diets. After fifty days (in May, 1941), the capillary resistance of all of these children was meas- ured again. Originally, there were sixty children in each group, but some of them had to be discarded because of intercurrent infections (mostly chicken- JreaUd group Con hoi oro up. Fig. 1. Effect of ascorbic acid 011 capillary fragility. The circles represent the capillary-fragility classes: Nor/«. = normal, S./. = slightly increased fragility, and D./. = definitely increased fragility. The figures sho\\- the number of children in each class in the beginning and (in parentheses) at the end of the experiment. The changes are indicated by arrows. pox) which might have affected their capillary status, so that finally only fifty- four children in the treated group and fifty-nine in the control group remained. The distribution of these children among the capillary-fragility classes and the changes observed during the experiment are shown in figure 1 . From the figure it appears that there were numerous changes in both direc- tions. In the treated group the capillary resistance rose in eleven children and fell in six; in the control group the figures were thirteen and ten, respec- tively. The sum total of these changes seems to be a slight improvement in the capillary resistance of the treated children, whereas that of the control children remained essentially the same. 578 Ascorbic Acid Subnutrition For further treatment of the data, the capillary fragility was expressed in numerical values as follows: normal = o, slightly increased fragility = i, and definitely increased fragility =2; definitely decreased fragility ^—2. The changes during the experiment were, accordingly, expressed by values ranging from —2 to -\-2. The observed changes are shown in table 2. There is a small improvement in the group given ascorbic acid, but the change is too small to have statistical significance. TABLE 2 Changes of Capillary Fragility in all Children Change, capillary- Number of children fragility units Treated group Control group — ^ 4 7 37 5 I I — I =to + 1 + 2 12 6 4 Total 54 59 Average change — 0. i5±o. 10* 0. i5±o ±o.oo±o. I I* Difference .15* * Standard error. Now, one might be inclined to assume that the circumstance that the groups consisted of a large proportion of children having normal capillary resistances may have overshadowed any possible effects of the treatment on the "deficient" children. Therefore it seemed worth while to examine the results after ex- clusion of the normal children. When the changes taking place in the capillary status of the originally "deficient" children were segregated from the data presented in figure 1 and the data treated in a way similar to that shown in table 2, the following values were obtained: the average changes were —0.82 ± 0.2 1 (S.E.) fragility units in the treated group and —0.68 ± 0.15 units in the control group. Thus there was a substantial improvement in both groups. It seemed to be greater in the group treated with ascorbic acid, but the difference between the groups (0.14 ± 0.26 units) was again too small to be statistically significant. Consequently, the results of this experiment do not permit the conclusion that treatment with ascorbic acid had materially improved the capillary resistance. Discussion and Conclusions It is clear from the foregoing that there exists a correlation between the as- corbic acid content of the diet and capillary fragility. The latter may be con- sidered to be some kind of a measure of the former. The correlation is, however, Osmo Tiirpeinen 5 '79 so low that the capillary-fragility test can hardly be regarded as a good method for assessing the nutritional status with regard to ascorbic acid. When large groups are examined, it may lead to fairly satisfactory results, but in smaller groups, to say nothing about individuals, the results obtained must often be grossly erroneous. In keeping with this low correlation is the observation that treatment with ascorbic acid often fails to return the increased capillary fragility to normal and to prevent the normal resistance from falling. It is thus evident that ascorbic acid cannot be the sole factor— and hardly even the primary factor— controlling capillary resistance in apparently healthy human beings. What the other factors are cannot, of course, be said on the basis of the evidence presented here. In this connection it seems, however, proper to refer to the recent investigations on vitamin P. Whatever the much disputed role of this factor is in the experimental scurvy of the guinea pig, it cannot be denied that in human beings vitamin P has frequently been found efficacious in returning increased capillary fragility to normal and in checking a tendency to bleed.^' ""■ ■'■ "* The view^ that human scurvy often should be regarded as due to a combined lack of both ascorbic acid and vitamin P, of which the latter would be the chief factor controlling capillary fragility, may prove correct. In view of the existence of such capillary controlling factors other than ascorbic acid, which possibly play a significant role in the etiology of scurvy, the capillary-fragility test may still remain a useful means of detect- ing subclinical scurvy^ although it undoubtedly is a poor index of ascorbic acid sub nutrition. Summary The correlation between capillary fragility and the ascorbic acid content of the diet was studied in 616 apparently healthy persons. The correlation co- efficient was found to be: ?■ = —0.155 ±: 0.040 (S.E.). Although of a low degree, the correlation thus was certainly statistically significant. In an experiment with a group of fifty-four elementary school children, treatment with 100 mg. of ascorbic acid daily for fifty days seemed to improve capillary resistance. This improvement, however, was so irregular that no statistically significant difference in the capillary status could be found be- tween the treated group and the control group at the end of the experimental period. It was concluded that, in general, increased capillary fragility is not a de- pendable sign of ascorbic acid subnutrition. It is evident that ascorbic acid cannot be the sole— and hardly even the primary— factor controlling capillary resistance in apparently healthy human beings. REFERENCES 1. Crandon, J. H.; Lund, C. C, and Dill, D. B.: New England Jl. Med. 223:353, 1940. 2. Goettsch, E.: Amer. Jl. Diseases Childr. 49:1441, 1935. 3. Scarborough, H.: Lancet 239:644, 1940. 58o Ascorbic Acid Subnutrition 4. Weld, C. B.: Jl. Pediat. 9:226, 1936. 5. Gothlin, G. F.: Skandinav. Arch. f. Physiol. 61:225, 1931. 6. Gothlin, G. F.: Lancet 233:703, 1937. 7. Hecht, A. F.: Jrbh. f. Kinderheilk., Ergzgshft. no. 113, 1907. 8. Jersild, T., and Elmby, A.: Klin. Wchnschr. 17:1359, 1938. 9. Ahlborg, N. G., and Brante, G.: Act. med. Scandinav. 104:527, 1940. 10. Gedda, K. O.: Skandinav. Arch. f. Physiol. 63:306, 1932. 11. Gothlin, G. F.: Act. paediat. 20:71, 1938. 12. Gothlin, G. F.; Frisell, E., and Riindqvist, N.: Act. med. Scandinav. 92:1, 1937. 13. Nordenmark, W.: Skandinav. Arch. f. Physiol. 70:186, 1934. 14. Dalldorf, G.: Amer. Jl. Diseases Childr. 46:794, 1933. 15. Molitch, M.: Jl. Laborat. & Clin. Med. 21:43, 1935. 16. Gianci, V.: Quadrni. nutriz. 6:527, 1940. 17. Sloan, R. A.: Jl. Laborat. & Clin. Med. 23:1015, 1938. 18. Abt, A. F.; Farmer, C. J., and Epstein, L M.: Jl. Pediat. 8:1, 1936. 19. Anderson, G. K.; Hawley, E. E., and Stephens, D. J.: Proc. Soc. Exper. Biol. & Med. 34:778, 1936. 20. Difs, H.: Act. med. Scandinav., supp. no. 110, 1940. 2i.Liebmann, J.; Wortis, H., and Wortis, E.: Amer. Jl. Med. Scis. 196:388, 1938. 22. OHara, P. H., and Hauck, H. M.: Jl. Nutrit. 12:413, 1936. 23. Rapaport, H. G.; Miller, S. H., and Sicular, A.: Jl. Pediat. 16:624, 1940- 24. Pearl, R.: Introduction to Medical Biometry and Statistics (3d ed.; Philadelphia and London: 1940). 25. Dalldorf, G., and Russell, H.: Jl. Amer. Med. Assn. 104:1701, 1935. 26. Armentano, L.; Bentsaht, A.; Beres, T.; Rusznyak, S., and Szent-Gyorgi, A.: Deutsch. med. Wchnschr. 62:1325, 1936. 27. Kugelmass, L N.: Jl. Amer. Med. Assn. 115:519, 1940. 28. Scarborough, H.: Biochem. Jl. 33:1400, 1939. THE EXPERIMENTAL PRODUCTION OF PSEUDOHERMAPHRODITISM IN THE MONKEY By G. VAN WAGENEN and JAMES B. HAMILTON From the DEPARTMENT OF OBSTETRICS AND GYNECOLOGY AND THE DEPARTMENT OF ANATOMY YALE UNIVERSITY SCHOOL OF MEDICINE NEW HAVEN, CONNECTICUT THE EXPERIMENTAL PRODUCTION OF PSEUDOHERMAPHRODITISM IN THE MONKEY* MODIFICATIONS of the urogenital tract in the genetically determined female have long been observed and described as occurring in man and domestic animals. Differentiation of accessory sex organs in a genetic female so that they resemble those of a male is termed female pseudohermaphroditism. Ob- viously, such profound changes in structure must have been initiated in early intra-uterine life. The masculinization of female offspring in mammals has been accomplished experimentally by Dantchakoff/ who used sex hormones to develop embryonic heterosexual genital ducts in the guinea pig, and by Greene and Ivy," who produced similar changes in the rat. In the experiments to be described in the present paper, masculinization of the female monkey fetus resulted in a striking resemblance to human in- fants who at birth have all the appearance of a male (cryptorchid), but whose subsequent developmental history is such that the question of sex does arise and in whom upon examination a uterus and ovaries are found. Material and Methods Testosterone propionate was the androgen used. Crystals dissolved in sesame oil in a concentration of 20 mg. per cc. were injected intramuscularly six times a week. The wish to spare the animal an abdominal section at the very begin- ning of the experiment led us to avoid the direct method of Dantchakoff, who introduced pellets of hormone into the amniotic cavity of guinea pig embryos. There were no previous data to indicate dosage in the monkey, so three graded dosages were arbitrarily chosen, 5, 10, and 20 mg. per day. Monkey 05 received 5 mg. per dose; monkeys 99 and 77, 10 mg.; and monkeys 96, 59 and 36 the largest amount, 20 mg. Table 1 summarizes for each individual the daily and total amounts of testosterone propionate administered and the time of the initiation and duration of the treatment. The rhesus monkey (Macaca mulatta) provides us with a primate laboratory animal in which anatomical and physiological characteristics of the repro- ductive organs are similar to those of man. For the purpose of properly timing the period for injection of the pregnant animal it was necessary in our study to have definite information concerning the beginning of pregnancy. The * This work was supported chiefly by grants to the Department of Obstetrics and Gyne- cology, from The John and Mary R. Markle Foundation and the Fhiid Research Fund of Vale University School of Medicine and also aided by a grant from the International Cancer Researclr Foundation to the Department of Anatomy. [583] 584 Hermaphroditism in the Monkey menstrual cycle of the rhesus monkey is normally about twenty-eight days long and the duration of gestation six lunar months (168 days). In our labora- tory the female animal is placed with the male for only forty-eight hours in a given cycle, so that the day of conception can be reckoned without great error. However, the early detection of pregnancy in the monkey presents greater difficulty than in the human because there is neither a "missed menstruation" to suggest that conception has taken place, nor a pregnancy test to confirm it. A menstrual cycle slightly longer than usual and a more prolonged bleeding TABLE 1 Hormone Administration to Mother Monkeys and Description of Fetuses Fetus No. Testosterone propionate given to mother Span of iniec- tion during fetal life," days Fetus Description of Fetuses Daily, mg. Total, mg. Age, days Weight, gm. Length, cm. 967 9 80 None c?68 None 963 960 0^66 c^86 20 20 20 10 10 5 1000 930 690 270 220 152 41 to 99 43 to 98 41 to 82 58 to 108 60 to 99 69 to 99 100 99 84 109 100 TOO 102 74 96 142 156 fibrou 200 macer 140 117 138 13-3 13-4 s mass 15.6 a ted 13-3 12.5 9-9 13-3 Female pseudohermaphrodite Female pseudohermaphrodite Male apparently unaffected Modified female Control Control Control at the time of the expected menstruation (placental sign) are two bits of evi- dence, neither one invariably present, which may indicate pregnancy. If rectal palpation then discloses that the uterus is gradually changing in consistency and shape, the diagnosis of pregnancy becomes possible sometime between the twenty-fifth and thirty-fifth day after conception. Experimental. Nine monkeys as they became pregnant were selected from the breeding colony. Most of these were parous animals which had delivered living young the preceding year. Six of these received a series of hormone in- jections beginning at various times after conception; the pregnancies of the other three were allowed to continue without treatment. Because of the relatively great length of pregnancy in the monkey, it was possible to save both time and the hormone by removing all fetuses as nearly as possible on the one hundredth day of gestation. As control material, a nor- mal female fetus was removed from monkey 72 on the one hundred second day, and a normal male fetus from monkey 94 on the ninety-sixth day. For a com- parative study, a younger stage in normal male development was sought and a seventy-four-day fetus removed from monkey 29 completed the control series. van Wagenen and Hamilton 58 K At the outset it was feared that an early injection of large amounts of hor- mone might cause abortion, and for this reason the animals first injected were given the smaller doses, beginning around the sixtieth day. When it was found that the hormone so given was well tolerated, higher doses were ad- ministered beginning about the fortieth day. Thus, the animals receiving the largest amounts of hormone were also those placed on earliest treatment. By analogy with human development, the genital organs at the fortieth day would still be in an indifferent stage with both male and female ducts present and the sex of the gonad unidentifiable by histological examination. Results. From the six androgen-treated animals, fotir living and developing fetuses were removed. In the other two monkeys the contents of the uterus were found to be degenerated. Fortunately, three of the foiu- living fetuses were genetic females and only one a male. Two were female complete pseudo- hermaphrodites. Pseudohermaphrodites are classified according to the sex of the gonad and the sexual type of the accessory sex organs. The first adjective applied to the pseudohermaphrodite is "male" if the gonads are testes, "female" if they are ovaries. The second adjective "external," "internal," or "complete" denotes whether it is the external, internal, or both external and internal genitalia which are at variance with the sex of the gonads. Thus, an individual with testes, whose external genitalia are female in type, is a male external pseudo- hermaphrodite. Monkeys 96 and 59, which received 20 mg. of testosterone propionate daily beginning on the forty-first and forty-third days, yielded female fetuses which have developed prostates and male external genitalia (fetuses 67 and 80, pi. 1, figs. 1, 5, and 6). The largest dosage of hormone (20 mg.) so completely changed the appearance of the fetuses removed from monkeys 96 and 59 that, as they were lifted from the uterus, the operator believed them to be normal males. The presence of hypospadias and the absence of palpable testes were the first hints that the fetuses might be, in reality, pseudohermaphrodites. The third female, fetus 63, came from monkey 05, which received a total dose of 152 mg. of testosterone propionate between the sixty-ninth and ninety- ninth days. Following this minimal amount and late treatment, the masculiniz- ing influence of the hormone was found in the enlarged clitoris and modified vestibule (pi. 1, fig. 4). Fetus 68 (pi. 1, fig. 2) was a genetic male from a pregnant animal injected with 270 mg. of hormone. No external changes were apparent unless possibly in the scrotal size. In monkey 36 vaginal bleeding and a dilated cervix led us to believe she was about to abort, in which case there was a danger that she might destroy the uterine contents. Monkeys usually eat only the placenta when a baby is born alive and vigorous, but may devour all the products of an abortion or an early parturition. In this monkey, hysterotomy was performed on the eighty-fourth 586 Hermaphroditism in the Monkey day of pregnancy and the fetus was found to have degenerated to a fibrous mass. Since pregnancy during the previous year had terminated with the birth of a normal baby, this deleterious result may have been the effect of the ad- ministered hormone. Monkey 77, which had not been pregnant previously, received 220 mg. of testosterone propionate between the sixtieth and the one hundredth day, a quantity less than that tolerated by monkeys 96 and 59. On the one hundredth day, the uterus was opened and found to contain a macerated fetus and placenta. Protocols Sinus hairs were present on the supraorbital areas and also circumorally in the fetuses of both treated and untreated animals. The remainder of the hairy coat, except that on the head, had not appeared in any fetus. Teeth do not appear until after term birth. Fetus 86, Untreated Control Male. The external genitalia were well formed. The anterior border of the scrotum, lying 2.6 cm. ventrocephalad to the ischial callosities, appeared as a turgid, partially bilobed swelling with a prominent median raphe, which was especially marked on the cephalic border of the scrotum from whence it extended to the under surface of the penis. The meatus was present as a sagittally placed slit. The prepuce was not distinct from the glans. The external appearance of this animal was similar to fetus 68 as shown in plate 1, figure 2. Sections of the testis with epididymis and selected regions of the reproduc- tive tract are shown in plate 2, figures 7 to 10, for comparison with androgen- treated male and female fetuses. The freely branching seminal vesicles were compressed to bluntly pyramidal structures. The lumen of each branch was circular in cross section, lined with cuboidal epithelium and occupied about one-third of the diameter. The lumen was somewhat narrow toward the ends of the branches and enlarged as it parsed backward to the stalk of the gland. The stalk lay in the connective-tissue mass with the vas deferens of the same side. Tissue of the cephalic lobe of the prostate was present dorsal to the ejacula- tory ducts and also ventrally between ducts and the urethra. This lobe of the prostate was loosely arranged, and isolated blunt ends of rudimentary glands protruded into the mass of connective tissue. The fibromuscular stroma of the cephalic lobe contained solid epithelial cores in which an occasional lumen had formed. Plate 2, figure 8, shows this part of the prostate. The utriculus masculinus can be seen between the relatively larger ejaculatory ducts. The caudal lobe of the prostate was the more compact and was situated on the dorsal wall of the urethra and for the most part below the entrance of the ejaculatory ducts into the urethra (pi. 2, fig. 9). The caudal lobe consisted chiefly of solid branching cords of epithelial cells with no lumen except in the ducts of the gland near their entrance to the urethra. van Wagenen and Hamilton 587 The bulbourethral glands were grossly twice the size of the homologous glands of the normal female fetus (compare pi, 2, fig. 10, and pi. 3, fig. 12). The epithelium lining the ducts was irregular and lumens were formed through- out. The excretory ducts of these glands coursed dorsal to the penile urethra and accompanied it for some distance into the spongy portion before opening into the dorsal wall. Plate 2, figure 7, presents the structure of the testis and epididymis. Fetus 60, Untreated Female Control. The gross appearance of the external genitalia of the animal is shown in plate 1, figure 3. The clitoris was a small non-protuberant mass, without a distinct glans, that overlay the anterior end of the vestibular orifice. The vestibule opened to the exterior between the anterior portions of the ischial callosities. The region over the symphysis pubis corresponding to the scrotum presented no swellings. The gonad was recognizable as an ovary with stroma and sex cords. Its character and that of the Fallopian tube and ampulla are shown in plate 3, figure II. There was a cavum uteri which in longitudinal section could be seen to dip into the uterine wall as irregular crypts, some of which were sub- divided (pi. 3, fig. 13). The epithelial lining of the uterus was composed of pseudostratified columnar cells, with their nuclei bordering upon the lumen, and the deeper parts of the cells appeared as a clear zone beneath the layer of crowded nuclei. The cells lining the crypts were cuboidal. The stroma, well developed and without edema, was composed of compact muscular tissue. The vagina lay dorsal to the urethra (pi. 3, fig. 12) to which in the distal portion it was bound by an outer common fibromuscular coat. At a site proximal to that in plate 3, figure 13, the cross section of the conjoint struc- tures was egg-shaped, with the vagina in the broad portion and the urethra in the narrow. The vagina was a solid cord of cells for the greater part of its length, but patent at the uterine end and also as it entered the vestibule. The vagina, whether a solid or an open structure, was compressed dorsoventrally throughout the greater part of its length, but just before opening to the sur- face, the vestibular part was compressed laterally (compare pi. 5, fig. 22, of animal 63 with pi. 3, fig. 14). Plate 3, figure 12, shows the urethra entering the vestibule on the summit of a papilla, while on the opposite wall of the vesti- bule there open ducts from the laterally located vulvovaginal glands. Essen- tially, then, the conformation of the vagina, urethra, and vestibule was that found postnatally in the normal female monkey. The clitoris was composed of two corpora cavernosa which in the peripheral part of this organ fused to form a single body. Nothing comparable to the corpus cavernosum urethrae was seen, but the tissue immediately surrounding the sagittal spike of cells attaching the frenulum to the glans was as loosely packed as that in the prepuce. Cavernous tissue was prominent only in the glans (pi. 3, fig. 15), although large veins were found within more proximal portions of the cavernous bodies. The clitoris had a core of non-ossified. 588 Hermaphroditism in the Monkey fibrous connective tissue in a position analogous to the structure forming the OS penis. The prepuce remained attached to the glans, both circumferentially and as a deep sagittal penetration on the under surface of the clitoris. There was no modification of the skin and underlying tissue in the region over the pubic bones. Fetus 68, Treated Genetic Male. The mother of this fetus received daily injections of 10 mg. of testosterone propionate from the fifty-eighth to the one hundred eighth day of gestation, a total dosage of 270 mg. The external genitalia of the fetus (pi. 1, fig. 2) are well developed, normal in appearance, and not appreciably enlarged. Plate 4, figures 16 to 20, are of histological sections comparable to those of the untreated control male, fetus 86, shown in plate 2, figures 7 to 10. The two lobes of the prostate gland and the bulbo-urethral glands of the treated male fetus were larger than those of the untreated male. The fibromuscular structure throughout the whole genital tract appeared better developed. In the injected fetus the pattern of the urethral lumen at the various levels and the contour of the surrounding parts were characteristic of the infant or young monkey. However, some of this increase in size and development must be dis- counted when it is noted that the control male fetus (96 days and 138 mg.) was younger and smaller than the injected fetus (109 days and 200 mg.). Histo- logically, the bulbo-urethral glands of the injected male apparently reflected androgenic stimulation. Changes in the prostate and seminal vesicles were less distinct. In the medullary portion of the testis of fetus 68 there is a longitudinal core of irregular, communicating spaces lined with a single layer of low cuboidal cells (pi. 4, fig. 16). The longitudinal section in plate 4, figure 16, can be compared with plate 2, figure 7, a cross section of the testis of the untreated male, fetus 86. The epididymal tube was lined with a single layer of cuboidal cells with large nuclei and comparatively little cytoplasm. The histological character of the epithelium was similar to that in the epididymis of the untreated fetus. Fetus 6^, Partially Modified Female (pi. 1, fig. 4). The treatment of the mother of this animal was deferred until the sixty-ninth day of pregnancy and then continued through the ninety-ninth day. Also, the amount of testosterone propionate employed was relatively small, 5 mg. per day with a total dosage of 152 mg. The vestibular orifice lay between the ischial callosities in the normal location for the female, and was surmounted by a clitoris, also in the normal position. However, the edges of the lateral walls of the vestibular orifice were irregular, clearly indicating a defect in their formation. The clitoris protruded for a distance of several millimeters and possessed a distinct glans covered only partly by the frenulum. Bilobed "scrotal swellings" were developed over the symphysis pubis. van Wagenen and Hamilton 589 The gonads were ovaries and contained no testicular tissues. The uterus re- sembled that of the control animal (compare pi. 5, fig. 21, and pi. 3, fig. 13). The vagina (pi. 5, figs. 22 and 23) was imperfectly formed. It consisted of two distinct dorsoventrally flattened tubes that from respective openings in the cervix and vestibule approached one another to overlap and end blindly. The overlapping extended for 1.6 mm. at the level of the neck of the bladder and in this region the two vaginal canals intercommunicated (pi. 5, fig. 23). Each part of the vagina possessed a separate muscular coat except in this region. The portion of the vagina opening to the exterior was the more ventrally placed, lying between the uterine portion of the vagina and the urethra. The two parts considered as a single vaginal canal resembled the vagina of the control animal in that a lumen was present only in the inner and outer thirds of its length. The lumen was lined with pseudostratified columnar epithelium for the most part, but the type of epithelium was variable with occasional patches of cuboidal cells. The nuclei were oval in shape and in the taller cells tended to lie in the middle third. Seminal vesicles and prostatic tissue were not observed. The bulbo-urethral glands lay ventral to the urethra as in the male, thereby differing from the position of the glands ventral to the vagina in the normal female. Tissues surrounding the vestibule had cavernous spaces, which in the parts lateral and ventral to the vestibule were especially noticeable (pi. 5, fig. 24). The clitoris, including cavernous tissues, was well developed. The homologue of the OS penis was present in the form of an elongated column of fibrous tissue (pi. 5, fig. 25). There was a distinct modification of the skin and underlying tissues com- posing the sex swellings. The epidermis was thin, with the stratum inter- medium lacking or reduced to a single layer. In this region the subepidermal tissue formed bilateral mounds of loosely arranged tissue of the same char- acter as that in the scrotum of the normal male fetus. The transition between the compact tissue of the corium of the normal skin and the scattered cells of the scrotal swellings was definite but not abrupt. Fetus 6y, Female Complete Pseudohermaphrodite. The mother of this genetic female received daily injections of 20 mg. of testosterone propionate from the forty-first to the ninety-ninth day of pregnancy. Structures, apparently a scrotum and penis, were developed at the location where normally these organs are found in males and quite removed from the usual position of the clitoris in females (pi. 1, fig. 5). The scrotum, like that of the control male, was large and bilobed with the phallus lying partially embedded between the lobes. There was no vestibular orifice in the perineimi but there was instead a prominent raphe in the midline extending from the anus to the urethral meatus. The phallus was hypospadiac, no gonadal mass was palpable in the scrotum or inguen. Internally the uterus and adnexa were similar to those shown in plate 7, fig- 590 Hermaphroditism in the Monkey lire 31, for the other pseudohermaphrodite, fetus 80. The ovaries, Fallopian tubes, and uterus of the two pseudohermaphrodites were the same size and occupied the same position as that found in the normal fetus. The vagina was short and passed ventrally, surrounded by prostatic tissue, to enter through the dorsal wall of the urethra below the neck of the bladder. The urethral end of the vagina was a flattened cord of cells, U-shaped in cross section. As it ap- proached the urogenital sinus it was reduced in size and greatly resembled the prostatic utricle of the male. A lumen was present at the cervical and urethral extremities of the vagina. Plate 6, figure 27, shows a section in this region which, except for the absence of the ejaculatory ducts, could be mistaken for that of a normal male. At a slightly higher level, part of the vagina on the left side was detached as an expanded vesicle. Prostatic tissue resembled in amount and degree of development that of the normal male of this age (compare pi. 6, fig. 27, and pi. 2, fig. 9). Glandular tissue was present as solid epithelial cords, very few of which had lumens. Open ducts passed from this tissue to enter the urethra both cranially and caudally to the level of entrance of the vagina. Seminal vesicles and ejaculatory ducts were absent. The bulbo-urethral glands (pi. 6, fig. 26) were about half again as large as in the normal female fetus with the proportion of epthelial tissue approxi- mately twice as great and the stroma reduced in amount. The gland was sim- ilar to that of the normal male (compare pi. 6, fig. 26, and pi. 2, fig. 10) but smaller than that of the treated male (pi. 4, fig. 19). The urethra at the level of the bulbourethral glands had abundant cavern- ous tissue (pi. 6, fig. 26). More distally in the penile portion of the urethra there were well-developed corpora cavernosa which lay above the roof of the urethra. Sections through the clitoris internal to the hypospadiac portion demonstrated a large amount of cavernous tissue and the presence of an os clitoris in which some bone had already been formed (see pi. 6, fig. 28). The prepuce was not separated from the glans. Fetus 80, Female Complete Pseudohermaphrodite (pi. 1, figs. 1 and 6, pi. 7, figs. 30 to 35). The mother received 20 mg. of testosterone propionate daily from the forty-third to the ninety-eighth day of pregnancy. Presumably this fetus and the preceding one were subjected to comparable treatment. The daily amount of hormone, the span of injection, and the weight of the mothers were very nearly the same. The two pseudohermaphroditic fetuses were similar except for the configuration of the scrotum and phallus, the external genitalia of this animal, fetus 80, being more like that of the male. The phallus had reached a position anterior to the scrotum, the two parts of which were united. In the other pseudohermaphrodite the phallus was short and occupied a posi- tion between the two lobes of the scrotum and the distinct raphe suggested fused lips of the labia (compare pi. 1, figs. 5 and 6). In fetus 80 the male ap- pearance was attained save for the hypospadias and cryptorchidism. van Wagenen and Hamilton 591 In both fetuses the scrotum was turgid and iranshicent. In fetus 80 the end of the penis projected slightly beyond the cephalic edge of the scrotum. The prepuce had not separated from the glans. The microscopical appearance of the ovaries, Fallopian tubes, and uterus was in all essential respects like that of the female control, whereas the vagina in terminating at the urogenital sinus was like that of the other pseudo- hermaphrodite. Plate 7, figure 33, presents the most distal part of the vagina as it enters the urethra, and at this level through the caudal lobe of the prostate the arrangement of all the tissues is typically that of the male. The tissue of the coUiculus seminalis is shown at a greater magnification in plate 7, figure 34. The scrotum was histologically identical with that of the normal male (pi. 1' %• 35)- Discussion The type of masculinization produced experimentally in these female mon- keys strongly resembled female external pseudohermaphroditism in humans, such as is observed in the adreno-genital syndrome. Sex reversal did not occur. Organs derived from the Miillerian ducts differ little in structure and position from those of the control female fetus. Structures derived from the Wolffian chicts, such as the vas deferens, ejaculatory ducts, and seminal vesicles, were not developed as a result of the treatment employed in these experiments. However, the hormone was not administered in any instance initil the end of the sixth week, when the gonads may have been differentiated into ovaries. This is probable since, in humans with their longer period of gestation, the sexual character of the gonads is already morphologically identifiable at this time. In the human female, glands corresponding to the prostate are found in the fetus lying above the Miillerian tubercle and may remain as rudimentary structures in the normal adult (Johnson'). Such structiuxs have not been ob- served in normal adult female monkeys and were not present in the control female fetus of this series. Following androgenic stimulation the prostate glands of the female monkey fetuses developed to fidly the size of those in the male fetuses. But in the human female pseudohermaphrodite the stimu- lated prostatic tissue is said to consist only of outgrowths from that portion of the urethra lying above the homologue of the vagina masculina or utricle in the male (Young*). This is in sharp contrast to the growth of prostatic glands in the masculinized monkey which arise both cranial and caudal to the Miillerian tubercle. The nipple of the mammary gland was present in experimental as well as control animals. In species such as mice and rats, in which the males do not have nipples, females subjected to androgenic stimulation fail to develop nipples (Hamilton and Gardner^). This suppression did not occur in the androgen-treated female monkey fetus. pq < H M Q U H , c CO s s^ ^1 n " (U ^ ^ (U ^ ■*- C/3 . CO « _ . ^ ^ c< '2 ' ^ r^ Oh " a. 2i r^-^ (Ll ^ (LI — _ 0, ^- 00 0 „ ■i~i 4- J-J pH 4- 4J U c c c >. c CO c "rt ON OJ -1- 0 ^ -5 CO r^ ^ 00" £1 °= t; ii - CL, ,, — s OJ CO -^ 00 0 o 13 ^- <^ ^ 2 3 l> '5 CO OJ 5 "-5 ■-5 o,^ (U C3 1- OS CO S2 o o Normal Normal Norm n 1 26 Absent Present 26 Absent Absent lale typ 27 Present (hyposp 0\ CI -T3 ^ <= CO tJ ■5 a! 1) Oh CO X 'C "^ .—V 0 CT c: K I- -0 M 4-; to w E E ^ E y: CI y: •1- c CI -0 ^ -5 o c C c c/" < 1 s 1 c/: rl ^ 0 ■— ' s "c « ,_^ -5 2 " E *« E cs E « E C-; E -i- c 4- C (L 4- c CO W-j c= il i- -H 1- *-< 1" ^ 1- c/- J: • rj- '^^ -i:^ o c c c dj C _C _c ;j^ r^ Z z z fe Z < a. < u Jj a c cj tiO 'C u 4- o 3 c/- -G C rt w rt C|0 a. 4-^ cr 4- bj D c 2 a. c "tr OJ > "2 c c a. 0. > C 4- 3 £ cr OJ > c * ^ ^ E 5 ■4- cr c C 0 6 5 4^ 2 _3 co" '4-' CO > 4- "cd r3 tL L. d. _ ^ c_ -c; (U c ^ U- > > D- c/: ix m CAl CLi H 1 van Wagenen and Hamilton 593 Summary Genetic female monkey fetuses, the mothers of which had been treated with androgenic hormone, developed as pseudohermaphrodites and exhibited anomalies which closely resemble those observed in human female pseudo- hermaphroditism (table 2). Six pregnant monkeys received daily injections of 5 to 20 mg. of testosterone propionate beginning upon the forty-first to fifty-ninth days of gestation and ending about the one hundredth day. Fetuses were removed from these ani- mals at the end of the period of treatment. Two males and one female fetus, removed from untreated mothers at a similar time in gestation, were studied as control specimens. The series of fetuses from mothers which had received hormone included three genetic females, two of which were pseudohermaphrodites and one which was less extensively modified by the treatment. One fetus was a genetic male which showed relatively little stimulation. The female pseudohermaphrodites had external genitalia that were male in appearance with a penis-like phallus containing the urethra lying cephalad to a well-formed scrotum. Internally, the tuerus, uterine tubes, and upper portion of the vagina were normal in structure and position. The vagina ended in the dorsal wall of the prostatic urethra and bore a striking re- semblance to a prostatic utricle. Prostatic glands were present and as large as those of the control male; they surrounded the vagina and had developed as outgrowths of the urogenital sinus, both above and below the Miillerian tubercle. Conclusion An exogenous androgen can pass through the placenta to induce pseudo- hermaphroditism in the female young of the monkey. Grateful acknowledgment is made to Miss Natalie MacCarthy, IVIr. Frank Caruso, and Mr. Joseph Negri who assist in the research of the Department of Obstetrics and Gynecology. Testosterone propionate was furnished under the trade name Perandren by Ciba Pharmaceutical Products, Inc., and Oreton by the Shering Corpora- tion. REFERENCES 1. DantchakofT, \'.: Cpts. id. Soc. de biol. 123:873, 1936. 2. Greene. R. R., and Ivy, A. C: Science (n.s.) 86:200, 1937. 3. Johnson, F. P.: Jl. Urol. 8:13, 1922. 4. Young, H. H.: Genital Abnormalities, Hermaphroditism and Related Adrenal Diseases (Baltimore: 1937). 5. Hamilton, f. B., and Gardner, W . I'.: Proc. Soc. Exper. Biol, c^- Med. 37:370, 1937. EXPLANATION OF PLATES PLATE 1 Fig.i. Monkey fetus 80: dissection to show anatomy of the urogenital tract and both internal and external changes present in the female complete pseudo- hermaphrodite; ovaries and uterus in position normal for the female; vagina entering dorsal wall of urethra well below bladder through prostatic tissue; the extended urethra of the male and scrotum of size and development character- istic of the male monkey fetus of this age except for the slight hypospadias. Fig. 2. Fetus 68: testosterone propionate-treated male, 109th day of gestation. Normal male appearance. Fig. 3. Fetus 60: control female, losd day of gestation. Fig. 4. Fetus 63: modified female, 96th day of gestation. Mother received daily 5 mg. of testosterone propionate (total, 152 mg.). Note length of clitoris and defective vaginal lips. Fig. 5. Fetus 67: female complete pseudohermaphrodite, looth day of gesta- tion. Mother received daily 20 mg. of testosterone propionate (total, 1,000 mg.)- Phallus does not lie in true male position in relation to the scrotum. Fig. 6. Fetus 80: female complete pseudohermaphrodite, 99th day of gestation. Mother received daily 20 mg. of testosterone propionate (total, 930 mg.). Com- pare with genetic male in figure 2. [594] Plate 1 [595 2 PLATE 2 All magnifications of photomicrographs are X 25, except figure 34 which is X 170. Fig. 7. Fetus 86: (normal male, figs. 7-10), testis and epididymis. Fig. 8. Fetus 86: urethra and paired ejaculatory ducts, the capsule of the latter including the prostatic utricle. Partially developed prostatic glands are present. Fig. 9. Fetus 86: inethra Avith seminal colliculus on the summit of which opens the prostatic utricle and laterally the open ejaculatory ducts. Many prostatic glands and ducts of the caudal lobe are also present. Fig. 10. Fetus 86: urethra with bulbo-urethral glands. 1:596] "^^ ^..*' '■» '^ ^$?^¥ "r.^m "^ v^>^v^ ■*;;» , •a dt ^b ^V^ ^L K «> ■•?i,??' ■^«5 10 Plate 2 C597] PLATE 3 Fig. 11. Fetus 60: (untreated female, figs. 1 1-15), ovary and adnexa. Fig. 12. Fetus 60: urethra and vagina. Bulbo-urelhral glands lie laterally. Fig. 13. Fetus 60: longitudinal section of the uterus. Fig. 14. Fetus 60: peripheral portion of vestibule. Fig. 15. Fetus 60: glans clitoris and frenulum. [5983 .'y-. ."i '. i %v ^/ // \\/ %7 ^\\ H 11 12 ^ m m ■# f"^' /■' 1 / 15 '*' V i' ^^ /^ ■7 Plate 3 PLATE 4 Fig. i6. Fetus 68: (treated male, figs. 16-20), longitudinal section through testis and head of epididymis. Fig. 17. Fetus 68: urethra and paired ejaculatory ducts, the capsule of the latter including the prostatic utricle. Partially developed prostatic glands are present. Fig. 18. Fetus 68: urethra with seminal colliculus on the summit of which opens the prostatic utricle and laterally open the ejaculatory ducts. Many pros- tatic glands and ducts of the caudal lobe are present. Fig. 19. Fetus 68: urethra with bulbo-urethral glands. Fig. 20. Fetus 68: glans jjenis illustrating lack of hypospadias. [600:1 »-f Jp * 1* ^ \t i ^t >0 -. '1 *><>.' PLATE 5 Fig. 21. Fetus 63: (androgen-treated slightly modified female, figs. 21-24) longitudinal section of uterus. Fig. 22. Fetus 63: urethra and two vaginal canals. Fig. 23. Fetus 63: intercommunication between two vaginal canals. Fig. 24. Fetus 63: peripheral portion of vestibule. Fig. 25. Fetus 63: glans clitoris and frenulum. [602] 21 [ 6o3 ] PLATE 6 Fig. 26. Fetus 67: (androgen-treated female complete pseudohermaphrodite, figs. 26-29), urethra with bulbo-urethral gland. Fig. 27. Fetus 67: section through prostatic gland showing vagina (prostatic utricle below urethra). Fig. 28. Fetus 67: glans clitoris proximal to hypospadiac portion. Fig. 29. Fetus 67: glans clitoris showing hypospadias. [604] 26 ^^.'' -•4-- ■v''*^-' > •^*. 27 ife> /. ■■^^"^pf \ \\J^^. 28 29 Plate 6 [605] PLATE 7 Fig. 30. Felus 80: (androgen-treated female complete pseudohermaphrodite, figs. 30-35), ovary and adnexa. Fig. 31. Fetus 80: longitudinal section of uterus. Fig. 32. Fetus 80: urethra and vagina, the latter surrounded by prostatic tissue. Fig. 33. Fetus 80: urethra with structure similar to seminal colliculus on the summit of which opens the terminal portion of the Miillerian ducts (vagina, prostatic utricle). Fig. 34. Fetus 80: higher magnification of colliculus from area shoAvn in fig. 33. Fig. 35. Fetus 80: scrotal swellings with median raphe. n6o6: 34 « ' »-* 33 '.«5r '35 Plate [607: THE HEART IN MYXEDEMA By JAMES J. WARING, M.D. From the UNIVERSITY OF COLORADO SCHOOL OF MEDICINE AND HOSPITALS DENVER, COLORADO THE HEART IN MYXEDEMA THE ASTONISHING iiumber and variety of the signs and symptoms of myx- edema (table i), their intriguing nature, the possibihties of accurate diagnosis of a malady that frequently masquerades as pernicious anemia or cardiorenal disorder, the remarkable mental, spiritual, and physical re- juvenation wrought by thyroid therapy have all contrived to make myxedema a disease fascinating to the clinician. This paper is a review of previously published observations on the changes in the heart in myxedema and a report of a case with enormous pericardial effusion. Experimental work on thyroidectomized animals as well as many clinical observations on myxedema in human beings all indicate the early develop- ment of cardiovascular disease of degenerative nature in the absence of the normal secretion of the thyroid gland. Evidences of cardiovascular disease consist in the main of arteriosclerosis, interstitial edema of the myocardium, swollen or vacuolated heart-muscle cells, and pericardial effusions. Around H. Zondek's paper of 1918^ on "Myxodemherz," giving details of four cases of heart failure due to thyroid deficiency and cured by thyroid medication, has arisen a lively discussion on the heart in myxedema. Zondek said all chambers of the heart were dilated, the heart action under the fluoro- scope w^as indolent, the pulse slow, and the blood presstae normal. In the electrocardiogram he noted marked diminution in the P and T waves. After treatment with thyroid he observed striking reduction in size of the heart, more vigorous action of the heart under the fluoroscope, a more rapid pulse without change in the blood pressure. In the electrocardiogram he observed a gradual return of the P and T waves to normal. In 1919 Zondek" said the aorta was dilated as well as both sides of the heart. In the electrocardiogram, P and T waves were greatly diminished or absent from normal cardiac cycles but were present in association with extrasystolic complexes. Under treatment the heart returned to normal size but the aorta still remained dilated. If in the course of treatment with thyroid the waves of the electrocardiogram become abnormally high ("Basedowkardiogram"), treatment should be reduced or stopped altogether. Finally in certain forms of myxedema, the cardiovascular- system complex may suggest the correct diagnosis. Although Assmann,^ Meissner,* and Zins and Rosier' in Germany and Fahr" in the United States published cases in substantial agreement with Zondek, others doubted if a cardiac abnormality existed sufficiently character- istic of myxedema to justify the term "myxedema heart." Fahr believes that heart failure is a prominent feature of an outspoken case of myxedema and furthermore that a mild degree of heart failure with very slight dilation of the heart is not at all infrequent in women patients past forty-five years ex- hibiting slight but definite evidence of hypothyroidism. Objective as well as subjective signs of heart failure, which do not respond to digitalis but are 6i 2 The Heart in Myxedema cured by thyroid therapy, may be present for many years. Fahr describes the "myxedema heart" as follows: All chambers of the heart are dilated, some- times enormously; the T wave in lead I is absent or negative; all cases show negative Q-R-S complexes in lead III during the stage of marked cardiac dila- tation; during thyroid medication, the dilatation of the heart disappears rap- idly and the T wave becomes positive. During the transition from a negative to a positive T wave, the T wave may be temporarily diphasic. A few cases TABLE 1* Myxedema: Symptoms and Signs I. General V. Circulatory 1 . Fatigue i . Shortness of breath 2. Weakness 2. Slow pulse 3. Intolerance to cold 3. Low pulse pressure 4. Subnormal temperature 4. Decreased minute volume, cardiac 5. Defective temperature regulation output 6. Increase in weight 5. Decreased blood volume 7. Increase in water content of the body 6. Increase in heart size 8. Serous effusions 7. Poor quality heart sounds 9. Local "tumefactions" 8. EKG changes 10. Slow speech 9. Feeble heart sounds 1 1. Deep, monotonous voice 10. Feeble heart movements 12. Joint pains and muscle stiffness ^,r ^^ 11 ■' 1 VI. Neurological 13. Intolerance to morphine and disjitalis t-> • -. 1 -i-^ ^ 10 1. Drowsniess, excitability II. Skin 2. Mental disorders 1. Pale, cold, dry, scaly 3. Slowness of thought and movement 2. Inelastic 4. Numbness and tingling 3. Absence of sweating 5. Impairment of memory 4. Hair, thin, coarse 6. Deafness 5. Brittle nails 7. Prolongation of tendon reflexes ^„ p. J 8. Electroencephalogram T, ,.j q. Diminution of tone in autonomic 1. Eyelids -^ r J u 1 system 2. Face and cheeks ^ . ,, 10. Increased protein in spinal fluid 4. Turgescence, infiltration of mucous VII. Gastrointestinal membranes 1. Poor appetite TTT Ti . 1 • 2. Achlorhydria IV. Hematologic ' T J u 1 . 1 ^- Indisjestion 1. Increased cholesterol ^ =■ . ■ 4. Constipation 2. Anemia ^ ^ „ , . . K. Hemorrhoids 3. Hyperproteinemia -^ 4. Decline in blood iodine VIII. Bladder atony IX. Amenorrhea, menorrhagia, metrorrhagia X. Low BMR XL Apparent absence of thyroid XII. Improvement with thyroid therapy * Modified from a similar table of Means. James J. Waring 613 show the split and prolonged Q-R-S complex of delayed intraventricular con- duction, which also appears under thyroid medication. Since the negative Q-R-S group in lead III, thought by many to be charac- teristic of left ventricular hypertrophy, becomes positive in these cases after thyroid medication, Fahr concluded that a negative Q-R-S group in lead III is not due to a preponderance of musculature of the left ventricle. In 1927, Fahr" reiterated his conviction that heart failure, as evidenced by dilatation of the heart, dyspnea on exertion, anasarca of the dependent por- tion of the body, passive congestion of the liver, and edema of the lungs, is not very infrequent in myxedema. If relief of these symptoms by thyroid ex- tract in amount sufficient to bring the basal metabolism up to normal and to cause the myxedema to disappear is evidence that both the myxedema and the heart failure have the same cause, namely thyroid deficiency, then the term "myxedema heart" is justified. Christian^ in his earliest paper (1925) did not note the changes so clearly described by Zondek. In 1928^ he said that digitalis was without effect in a few patients with enlarged heart and signs of congestive failure due to myxedema but that thyroid medication resulted in marked improvement. In 1935'" he thought pericardial effusions were probably present in those instances in which gieat apparent enlargement of the heart markedly and rapidly de- creased under thyroid therapy. In other instances atony as well as edema of the heart muscle played a part in producing enlargement. Of 162 cases of severe myxedema studied with special reference to the cardio- vascular system by Willius and Haines," 148 or 90 per cent showed no sub- jective symptoms or objective evidence of organic cardiovascular disease. Eleven or 7 per cent showed definite signs of organic cardiovascular disease which was, in only one instance, appreciably influenced by the disappearance of myxedematous symptoms under thyroid treatment. Analysis of X rays of the hearts in this series was not given. Willius and Haines conclude that their observations do not justify the establishment of a cardiac syndrome charac- teristic of myxedema. Christian^ warned against the possibility of circulatory failure from the incautious use of thyroid medication in myxedema. Although some cardio- circulatory disturbance may be due to hypothyroidism and correctable in the earlier stages, the later stages of circulatory disturbance will not be changed because of their permanent organic nature. In intermediate stages a properly balanced treatment with digitalis and thyroid substance is indicated and may produce good results, provided the thyroid dosage does not overburden the heart. In forty-eight myxedematous patients seen at the thyroid clinic of the Massa- chusetts General Hospital in the preceding twelve years, Means, White and Krantz"^ found only one in which there was a definite cardiac enlargement which subsided under thyroid. They concluded that "myxedema heart" in the sense of Zondek and Fahr, is far from common. On the other hand they 6 14 The Heart in Myxedema were not prepared to say that it does not exist. When symptoms and signs o£ heart faikire are present, a coexisting hypertension and arteriosclerosis will be found. In 1933, Lerman, Clark and Means^^ summarized their subsequent observa- tions as follows: "Myxedema heart" in the sense of cardiac enlargement which undergoes shrinkage on thyroid medication is common. In the sense of Zondek and Fahr, that is, a cardiac enlargement in association with congestive failure, it is rare. The heart changes are part of the picture of myxedema and not a separate cardiac entity. Patients with cardiac failure usually have hypertension and arteriosclerosis. The change in the heart size with thyroid treatment is probably due to increased muscle tone and loss of interstitial edema, occa- sionally to loss of an accumulation of pericardial fluid. In 1933, Means" suggested that the cardiac enlargement found in some cases of myxedema may be due either to hypertrophy, dilatation, or edema of the heart muscle. Pericardial effusion may be possible but the X-ray appearance is against it. In further studies on the heart in myxedema, Lerman, Clark and Means^' state that all of eighteen patients showed enlargement of the heart which de- creased in size after thyroid medication. None showed congestive failure. All had abnormal electrocardiographic changes. Subsequently, Means^" described the state of the heart in myxedema in brief, as due to dilatation and low tonus, but without true insufficiency, since along with the reduction in the heart's capacity for work there goes a parallel reduc- tion in the amount of work it is called upon to perform. "The condition of the heart in myxedema is not a complication, it is part of the disease, myxe- dema of the heart in fact, and whether it is to be looked at as in a state of failure is, to a certain extent, a matter of terminology." As to digitalis, he says, myxe- dematous patients tolerate digitalis badly just as they do morphine. In 1932, Fahr" summarized his observations on the heart in seventeen cases of severe and moderately severe myxedema. In thirteen instances or 75 per cent he found signs and symptoms of heart failure, all of which disappeared after giving thyroid. If coronary disease is present he warns that the blood supply to the heart in proportion to the work done may be much less after than before giving thyroid. Provided the heart shows better function and provided no angina pectoris is induced, thyroid may be given. If the coronaries can dilate, anginal attacks may disappear after thyroid treatment is started in myxedema. Sturgis and Whiting" reported the sudden death of a myxedematous patient shortly after treatment with thyroid was started. They urge caution in thyroid therapy, especially in myxedematous patients with anemia and recommend preliminary treatment with small transfusions and digitalis. Ohler and Abramson^^ report the sudden death of a young myxedematous patient given enormous doses of desiccated thyroid. Less serious unfortunate effects of thyroid therapy, auricular fibrillation (Swan^°) and Stokes-Adams James J . Waring 6 1 5 attacks (Thompson"'), in both instances disappearing when thyroid medica- tion was temporarily withdrawn, have been reported. On the other hand, the pain of angina in myxedema may disappear dining thyroid therapy, as in the case reported by Ziskin." The blood pressure in myxedema may be low, nor- mal, or moderately elevated. Under treatment the general tendency is for the jjressme to move toward a normal range (Ohler and Ullian"''). In a valuable study of thirty-five cases of myxedema seen at the Boston City Hospital in the preceding seven years Ohler and Abramson'" found character- istic changes in the electrocardiograms in thirteen, namely decrease in voltage in all the complexes and frequently an inversion of the T waves in all leads. The T waves in lead I were abnormal in every one of these thirteen. Also the P curves were low in all but one case. Prolonged A-V time was occasionally seen and convex S-T intervals in three cases. With few exceptions these changes were observed when the basal metabolic rate dropped to a level of —25 or lower. In some cases under treatment it was noticed that the electrocardio- graphic changes returned to normal before the basal metabolic rate had risen to normal. Enlargement of the heart occurred in seven of the thirteen cases showing electrocardiographic changes. With one exception, perhaps due to the brief period of treatment, the increased size of the heart was reduced as a residt of thyroid therapy. In several instances, therefore, the heart was not enlarged and yet the electrocardiogram was characteristic. Distant heart sounds and "mild congestive failure" were frequently found. They conclude: "The changes noted in this series are characteristic of the disease and warrant the use of the term, 'myxedema heart.' " Ohler and Abramson suggest that the presence of a myxedematous infiltra- tion of the heart, a mucinous tissue involving the muscidar fibers and inter- fibrillar spaces and nervous elements of the heart, which disappears with thyroid therapy, may in small quantity produce only changes in the electro- cardiogram without increase in size of the heart, but in larger quantities w^ill cause dilatation. They do not believe that resistance of the skin can cause in- version of the T waves nor prolongation of the conduction time. Coelho"* does not think cutaneous resistance important, since the use of needle electrodes gave the same results as ordinary electrodes. Paul White also by the use of needle electrodes showed that the altered conductivity of the myxedematous skin had no influence on the graphic records. Bellet and McMillan"" ascribe the electrocardiographic changes to alteration in the myocardium. The low voltage may be explained by easier dissipation of the heart current through the myxedematous tissue. They do not mention S-T displacements among the characteristic changes. Ohler and Abramson'^ call attention to the close resemblance between the variations in the electrocardiogram in certain cases of myxedema and the picture not infrequently seen in coronary sclerosis as originally described by Pardee. The low voltage, the inverted T waves, and the upward convexity of the S-T interval are common to both diseases. However, in their cases of 6i6 The Heart in Myxedema myxedema the amplitude was constantly low, while in coronary sclerosis the Q-R-S complexes though frequently low may be high. In the one condition the electrocardiogram returns to normal with thyroid therapy, in the other it remains unaffected. In only one of the thirty-five cases of myxedema was there a suspicion of pericardial effusion on fluoroscopic and roentgenographic ex- amination. The fact that in general the cardiac pulsations returned to normal under thyroid therapy with very little change in the size of the heart makes the presence of pericardial effusion in those cases at least unlikely. Pardee"" also says the S-T segment may show an upward convexity suggest- ing the coronary T wave but in contrast to the usual appearance of the coro- nary T the voltage of T in myxedema is small. The A-V time is frequently prolonged. These changes, he thinks, are due to physiological variations in the muscle metabolism and not to pathological changes, because the adminis- tration of thyroid brings about their reversal. Katz" thinks the electrocardiographic changes in myxedema are due to changes in the water and protein content of the heart and surrounding tissues brought about by removal of the normal action of thyroid secretion. The changes comprise (i) sinus bradycardia, (2) low Q-R-S complexes in the limb leads, (3) small or inverted T waves, and (4) S-T depressions. Gross changes in the heart include dilatation, sluggish pulsations, pericardial effusions. Coro- nary insufficiency may develop in either hyperthyroidism or hypothyroidism. In the latter case it is probably due to the combined effect of the frequently present anemia and interference with oxgen by the myxedematous cardiac muscle. As for large pericardial effusions in myxedema. Freeman,"* and Gordon" have reported such cases. In each instance the patient made satisfactory recov- ery with thyroid therapy. Suspicion naturally arises that the slight and the moderate "cardiac" enlargements seen so frequently in myxedema might be due to small pericardial effusions and not, as at first thought, to dilatation of the heart. Pericardial effusions are frequent in thyroidectomized animals, but in human myxedema their presence or absence can be settled only through many post-mortem examinations or through the hazardous aspiration of the "moderately enlarged hearts." Pathologists rarely have the opportunity to examine a person dead of untreated myxedema. Furthermore, myxedema is by no means a very common malady. In the following report of a case of enormous pericardial effusion in myxedema, the later changes in the size of the heart shadow were quite probably due to recurrences of the effusion. Case Report D. C, male, farmer; age 5^ years; myxedema; hies; enormous pericardial effu- sion; ascites; pleural effusion; atony bladder; co^nplete disappearance of peri- cardial effusion after ^ cc. Salyrgan intraveyiously; typical electrocardiographic changes; recovery xuith thyroid therapy. This man entered the Colorado General Hospital on January 27, 1935, with James J. Waring G 1 7 the complaints of marked constipation, increasing shortness of breath, "pain in the stomach," enlargement of the belly, and swelling of the feet. About six months previous to admission he first noted difficulty in getting his bowels to move. In spite of frequent cathartics, he was still troubled with constipation and some distress in the upper part of the belly, which had slowly but markedly increased in size for the previous five months. All this had been accompanied by slight swelling of the feet and ankles, shortness of breath, increasing weakness and lethargy. Past history was unimportant except for the fact that in igi6 he contracted syphilis for which he took very little treat- ment. Physical Examination. On admission the patient was comatose, answered questions not at all or only after a notable pause and then in a deep, monoto- nous voice and without opening his eyes. He seemed in no respiratory distress while lying flat in bed. The skin was dry, brown, scaly, not cyanotic; the mucous membranes of the lips were a little cyanotic. The hair was thin, espe- cially on the back of the neck and above the ears. The eyebrows w^ere thin in the outer thirds. The eyelashes were thin on the upper eyelids and missing from the lower lids. Pubic and axillary distributions of hair were normal. The face was generally swollen, the eyes narrowed to slits by edematous upper and lower eyelids. The belly was much distended, with dullness in the flanks, and the feet and legs were slightly edematous. The hands were dry, cold, thick, the skin over the backs parchmentlike. There was some occupational der- matitis. The tongue was dry, coated, thick; the mouth edentulous; there was slight cervical adenopathy; the thyroid gland was not palpable. Both pupils reacted sluggishly to light and diuung accommodation. Chest. The cardiac impulse was not visible nor could the movements of the heart be felt by palpation. On percussion the area of precordial dullness ex- tended from right anterior axillary line to left anterior axillary line at the levels of the fourth and fifth ribs. The interspaces over this area were filled out. The mediastinal dullness was about 14 cm. wide in the third interspace. The heart sounds were inaudible. The veins in the neck were somewhat promi- nent. The radial pulse was regular in rhythm, rate 60, temperature 97°, respira- tion 28, blood pressure 150/95. The lungs were resonant in the upper lateral portions of the chest and also in the upper portions posteriorly. The base of the left lung was dull. Signs of pulmonary edema were absent. Abdomen. The abdomen was distended, tympanitic superiorly and dull in the flanks. A fluid wave was present. The liver was just palpable. The spleen was not felt. Extremities. The lower extremities showed a slight edema. The knee jerks were present but sluggish, with a quick contraction and tardy recovery typical of myxedema (Chaney""). Genitalia. Penile scar present; prostate moderately large and firm. Laboratory Tests. Urine: specific gravity, 1.030. No albumin, sugar, acetone, or casts, but a small number of pus cells. James J. Waring 6 1 9 Blood. Hb. 11.2 gr.; red blood cells, 3,950,000; white blood cells, 9,850. Differential count: neutrophiles 67 per cent, lymphocytes 25 per cent, endo- thelial cells 7 per cent, eosinophiles 1 per cent. Wassermann test +4. Eagle Hocculation test, positive. Blood sugar 80, nonprotein nitrogen 14. On 1-28-35 catheterization showed 150 cc. residual urine. Utotropin gr. X t.i.d and acid sodium phosphate gr. XX t.i.d. were given until 2-4-35 '^vhen the urine became bloody. They were then discontinued. The urine resumed nor- mal appearance on 2-1 1-35. On 1-29-35 portable A-P X-ray film of chest showed immense enlargement of the cardiac silhouette (fig. 1 .) On the morning of 1-30-35, 560 cc. of straw-colored serous fluid was aspirated from the pericardial sac. No attempt was made to withdraw all the fluid. The blood pressure before aspiration was 160/95, ^fter aspiration 135/90. Heart sounds were now faintly audible. The area of cardiac dullness seemed but little aff^ected by the removal of the fluid (fig. 2). The Wassermann test on the pericardial fluid was +4. After aspiration, a portable A-P X-ray film of the chest showed very little reduction in size of the cardiac silhouette. Microscopic examination of the fluid removed from the pericardium: The sediment after centrifugation was highly cellular. It was composed chiefly of monocytes, a few small lymphocytes, and only an occasional polymorpho- nuclear leucocyte. A few acinar structures lined with cuboidal epithelium con- taining hyperchromatic granular nuclei were present. No acid-fast organisms or other bacteria were found by smear or culture. On 1-31-35 he was given 1 cc. Salyrgan intravenously and on 2-1-35 2 cc. Salyrgan intravenously. He was given also theobromin sodium salicylate gr. X t.i.d. every other day until 2-5-35. Meanwhile, the area of cardiac dullness rapidly diminished. A six-foot posteroanterior film (fig. 3) on 2-4-35 showed a cardiac silhouette of normal size, a cloudy left costophrenic angle, possibly due to a small amount of pleural effusion and a triangular area of platelike atelectasis or interlobar effusion in the right lung. The first electrocardiogram on 2-4-35 (^S- S) showed changes characteristic of myxedema. Notes on this and subsequent electrocardiograms are shown in table 2. On 2-9-35 '^^^ BMR was —40. Thyroid therapy was begun cautiously wath small doses of desiccated thyroid, i/^ gr. daily, increasing the dose until by 3-19-35 he was taking i/o S^- fo^^^ times daily. Antiluetic treatment consisted of daily inunctions with blue ointment, sodium iodide intravenously, and later potassium iodide by mouth. On 2-12-35, when the BMR was —40, tests for hearing defects showed partial perceptive deafness, which later improved under thyroid therapy. On 2-19-35 the total plasma proteins were 6.2, albumin 3.1, globulin 2.9, fibringogen 0.2. The albumin-globulin ratio was 1.07. On 2-22-35 a lumbar puncture showed the spinal fluid under normal pres- sure. The fluid was clear, contained very few cells, sugar 71 mg. and total 1 1 1 1 1 1 1 1 r-> ^« > 2 o -5 n w u o iM u Z r> t/: tb o c 5 g 4_) OJ o ^ O "mi CI u u-l i-i ^ o ^ u c« pl> o 'cA ^ rt u 2 OJ v OJ a r\ 13 n OJ < , il 3 "a, r-" '^ n o 5J 8 r- "rt ei 3 ^ ^ r^ c/5 E > ^J
  • o 0 0 00 >o 00 t^ oc 4 c 1 + el + CI + + + Cl + + Cl + + + 8 + .2 . «i .2 c P- 0 cc (5C 0 0 w-1 +J CO (JC c 0 0 0 -t- 00 CO 1 0 1 1 1 0 ? 0-, 1 1 d T 1 CI 1 CI CO 1 cl 1 cl C", 1 0 CO 1 CO V d ^. 1 1 ■ cl cl 1 1 0 CO w H-l m to Voluntarily stopped thyroid this day for next 3 months Note gain in weight and declining BMR Condition now typically myxedem- atous. Note gain in weight, low BMR, high cholesterol. Thy- roid resumed. Heart definitely larger by percussion. Heart sounds more faint C '> 0 £ C '> 2 a. > a .£ _>. ■*-* lU C .2 -5 c 0 U 4-1 G JJ 1j u X ^ o c D c c o -C CD a o CO o — _l c o f^l D U 00 q .0 I* • > CM \t- CO 1 . , .1 •5=0 '^ "^ I 1 1 1 1 1 • O O O O O O c O 00 CD ^ CM (J 634 Growth o£ Deer Antlers masses situated on either side of the base of the bladder, compactly ensheathed by connective tissue, and varying from 2 to 4 cm. in length and about 1 cm. in diameter, depending on their fvuictional condition. In the June and July ani- mals (pi. 2, figs. 5 and 6) the seminal vesicles are minimal in size; in the August animal (pi. 2, figs. 7 and 8) they are distinctly enlarged; while in the October specimens (pi. 3, figs, g and 10) they are about twice the size of the June glands. The histology also undergoes a marked change. In the June and July specimens (pi. 2, figs. 5 and 6) the gland presents numerous small cavities lined by rela- tively low, inactive cells. The connective-tissue partitions between the glandu- lar spaces are relatively thick. Very characteristic of these months is the presence in the lumina of darkly pigmented, amorphous masses which give the impression of being inspissated secretion. In August (pi. 2, figs. 7 and 8) the lumina are distinctly enlarged and the epithelial cells lining the mucous mem- brane are considerably taller than in June and July and show signs of secretory activity. The black, amorphous masses of secretion have diminished. By Oc- tober (pi. 3, figs. 9 and 10) the gland is markedly changed, its cavities having become very much enlarged and filled with copious amounts of globular and granular secretion. Accompanying this enlargement, the partitions separating the numerous cavities have become stretched, transforming the mucous mem- branes into an intricate system of trabeculae or septa. The epithelial cells have become exceedingly tall and slender, with the nuclei pushed to the bases of the cells. From an average height of around 18 to 20 /a in June and July the cells have attained a height of about 40 ^a. None of the black, amorphous, in- spissated inasses of the preceding stages are present any longer. From these observations it may be concluded that the seminal vesicles, as well as the testes, of the Virginia deer undergo pronounced seasonal variations. Both the testes and seminal vesicles which are in a state of regression in May and June become gradually activated, to reach a maximum state of activity in the fall, when rutting occurs. It is known that in other mammals the male ac- cessories are maintained by the male sex hormone produced by the testes; consequently, it seems fully justifiable to relate the seasonal activation of the seminal vesicles, shown here in the Virginia deer, to a seasonal activity of the testes. One may deduce that the testes of the Virginia deer begin to become active late in June, and that they in turn begin to stimulate the male acces- sories in July until these reach their maximum activity, together with the gonads, during the fall rutting season. The Effects of Castration on the Groivth and Periodicity of the Antlers. The periodicity of the gonads and seminal vesicles invites comparison with the antlers, a male secondary sex character which is also subject to seasonal changes. In a separate paper the writer" has described the annual cycle of the deciduous antlers of the Virginia deer, based upon a study of the same animals utilized in the present report. Briefly stated, the antlers of the white-tailed deer commence to grow late in April or early in May; they attain their full George B. Wislocki 635 growth, become hard, and shed the velvet in September, while in late De- cember, or in January, the antlers are cast off.* Examination of the schedule for gonads and antlers (graph i) re\eals that the antlers begin their annual renewal at a time when the testes and seminal vesicles are most inactive; they become hard and the velvet is shed when the testes and male accessories are rapidly enlarging, whereas the antlers are lost when the reproductive organs have begun to decline. These observations sug- gest that some nongonadal factor must be responsible for the initiation of antler grow^th. Fortunately for our knowledge of the antler-gonad relationship, numerous deer have been castrated at various times. Few records of these ex- periments, however, have been kept, beyond in some instances the antlers themselves, or pictures of them. The most extensive account of such experi- ments has been written by John Dean Caton,'- a judge in Illinois who studied the natural history of deer as a scientific avocation. He castrated numerous Vir- ginia deer and elk. From his classical account of these experiments the fol- lowing passages are excerpts. They illustrate with the utmost clarity how the antlers become typically deformed following castration. "If a deer be castrated at any time after the antlers are so far matured that their velvet may be removed without material injury, and while they still firmly occtipy their seat, they will irivariably drop off within thirty days there- after, though it may be months before the time that they would have been shed in the course of nature. In this case, and also when the operation is per- formed after the antlers are dropped naturally, in the spring following w'hen the new antlers on the perfect buck commence their growth, the same growth commences on the mutilated animal, and progresses to all external appearance the same as on the perfect animal till they have attained nearly the same size as those which were last cast off. If the buck be a young one with a spike antler, this will be a spike also of nearly the same length. If an old buck with five tines, these will be of nearly the same size as the former, with five tines also. These, however, never perfect their growth and never lose their velvet; but at the time the antlers on the perfect buck lose the velvet, those on the mutilated bucks stop their growth, but a moderate circulation is kept up in the velvet, which remains warm to the touch, and so they continue stationary till the severe weather of winter freezes the antlers through down to or very near the burr, when by the application of some accidental force they snap oft within half an inch or an inch of the burr, depending on the size of the antler. If w^e now examine the detached portion of an antler we shall see that * With variations of a montli or two, a similar calendar holds for the majority of deer in the temperate zone north of the equator, the only marked exception being the European roe deer (Capreolus capreolus). In this unusual species the antlers appear in January, and the breeding season occurs in August (Bischoff;' Keibel,* Stieve"). With this odd calendar the females combine the imique fcatiue among deer of displaying delayed implantation, namely, a period of three to four months before implantation occurs during which the fertilized eggs lie dormant in the uterus. Implanted, as in other deer, at the usual time (November), the young are consequently also born at the usual time (May). 636 Growth of Deer Antlers its entire body is loose and spongy, more condensed at the circumference than within, but has nowhere attained the consistency of hard bone, so as to close up the blood-vessels leading into it from the periosteum. "These stumps of the antlers are carried till the next spring, when a new antler shoots out from the old stump not so large as its predecessor, and grows on in the same way and at the same rate as on the perfect animal, till those so far mature as to shed their velvet, when as before that on the mutilated animal stops its growth. In the meantime the old stump has enlarged its diameter and put out large tubercles as if supplemental to the burr, which is also con- siderably enlarged. The new antler thus produced is not so large as the former, and if branched has less tines. And so this process goes on year after year, each succeeding antler being less in size and perfection than its predecessor, while the enlargement at the lower end becomes an exaggerated burr. This process of growth differs very considerably in different individuals of the same species. In some, in a few years, these stumps grow to an enormous size, covered all over with large tubercles, some of them amounting to shafts two or three inches long, which may be frozen and broken off in the winter, while neither may be so conspicuous as to be recognized as a beam. The whole of this irregu- lar mass is ever covered with the fine, soft, glossy fur. These two large masses in the place of the antlers, covered all over with these rudimentary shafts, present a very curious and interesting appearance on the head of a deer." Similar deformed antlers in various North American deer, following cas- tration, are mentioned or pictured by sketches in the writings of Seton." Town- send and Smith" show several photographs of two Virginia deer which had been castrated. The deformed antlers resulting from castration are sometimes referred to as "cactus-antlers" in the United States. Similar observations for European deer of various species have been col- lected by Gaskoin,^^ Fowler," Rorig,"'^* and Brehm.^" In the European roe deer a similar but more peculiar and striking deformity of the antlers occurs following castration. In these deer the antlers become converted into huge masses of rubbery, exuberant protuberances, which are permanently covered by fur, remain vascular, and are only moderately calcified. An outspoken example of this type of antler deformity, designated as "wig antlers," because of its resemblance to an old-fashioned wig, is shown in a photograph by Tan- dler and Grosz.^° A related, although somewhat different, response to castration has been observed in the reindeer (Tandler,"^ Tandler and Grosz'"). In rein- deer oxen, according to Tandler, the antlers are much larger than in the bulls, and they never lose their velvet completely. Reindeer oxen, according to these observers, contrary to all other observed species of castrated deer, do, however, shed and renew their antlers annually, although the antlers are shed several months later in the oxen than in the bulls. However, Jacobi" adds the ob- servation that the antlers of castrate reindeer, due to a preponderance of soft, spongy bone, are not so hard as those of normal animals. Contrary to George B. Wislocki 637 Tandlcr, Jacobi claims (p. 238), however, that totally castrated oxen drop their antlers at the same time as bulls. It should be remarked, however, that the data regarding castration in reindeer are not so trustworthy or complete as for other deer, because reindeer are frequently mutilated by the Lapps by crushing the testes, instead of by actual castration, and in the cases cited by Tandler and by Jacobi there is no evidence that the degree of castration was ever verified in any individual instance by post-mortem examination. Tandler and Grosz also contribute briefly the statement that in a doe of the European roe deer, as well as in a second doe of undesignated species, both of them belonging to species in which normally the females do not possess antlers, removal of the ovaries did not induce formation of antlers. Similarly in a female reindeer, a species in which the female normally possesses antlers, the annual growth and renewal of the antlers was not modified by bilateral ovariectomy. Various interpretations have been placed on the observations outlined above. Regarding his own experiments with deer, Caton'^ remarked in 1877, "We may admit that one physical body can only produce a physical effect upon another body by a physical medium, and so conclude that there must be a physical medium between the testes and the antlers, specially designed and qualified to produce the effect observed; but if so it is as yet not identified, and we can only hope that some more ingenious and careful observer may find it. The utmost we may safely say now is that in some way the testes enable or stimulate the proper blood vessels to carry into the antler a larger amount of earthy matter and there properly deposit it than they can do after the testes are removed, presuming at the same time that the absence of the generative organs deprives these vessels of, or weakens other important functions neces- sary to, the full maturity of the antler." Tandler and Grosz"^ summarize their observations on roe deer to the effect that castration at a time when the velvet is present leads directly to the forma- tion of wig antlers, while castration after the velvet has been shed imposes prompt loss of the antlers and subsequent development of wig antlers. Con- cerning reindeer Tandler' concludes that antler formation is independent of the presence of the gonads. A castrated reindeer, whether male or female, renews its antlers annually, as do the individuals who are in possession of gonads. Consequently, according to Tandler, the antlers of reindeer represent a species character which is independent of the gonads. A more recent contributor (Olt,"^), speculating on the gonad-antler relation- ship based on consideration of castrated roe deer, begins with the assumption that spermatogenesis is limited to the rutting season, whereas the testicular hormones are given off constantly throughout the year. Furthermore he postu- lates four testicular hormones: one, which induces antler growth; a second, which regulates growth; a third, which initiates resorption of the velvet by a marked local "eosinophilia"; and a fourth hormone, which acts on the muscu- 638 Growth of Deer Antlers lature of the blood vessels of the antlers inducing a nutritional disturbance, so that the antlers die and are eventually cast off. In a second article Olt"* concludes further that there is a seasonal inhibition of antler growth by the testes, expressed by the withering of the velvet and subsetjuent death of the antlers. The present observations on the annual cycle of gonads and seminal vesicles indicate that the basic assumption of Olt is false; contrary to his belief, the present evidence supports the conclusion that the internal secretion of the testes, which is responsible for the changes in the male accessories, in- creases and declines annually in a time sequence coinciding with the rise and fall of spermatogenic activity. The present writer would interpret the annual sequence of events in the normal antler-gonad cycle, viewed in the light of the results obtained by cas- tration, as demonstrating that nongonadal factors are responsible for the ini- tiation of antler growth in adult deer, but that full maturation of the antlers in all species of deer depends to some degree on the presence of intact gonads. This conditioning by the testes is greater in the Virginia deer and roebuck, however, than in the reindeer. Nevertheless, in the latter, contrary to the conclusion reached by Tandler and Grosz, an appreciable testicular effect is actually present, judged by observations made following castration (increased size, failure to harden completely, longer retention of velvet). In female deer, on the contrary, judging from the fragmentary observations of Tandler and Grosz, whether the female possesses antlers (reindeer) or not, gonadectomy produces no apparent change in regard to antler formation, and hence in the female the antlers or their rudiments are possibly unconditioned. From what has been observed in normal as well as castrate deer, it ap- pears that the antler cycle must be regulated by both nongonadal and gonadal factors. In the normal cycle of the white-tailed deer the nongonadal factor obviously initiates antler growth before the testes have revived from their involution of the previous winter, while in castrate deer this factor must be responsible for the continued formation of antler tissue. The gonadal factor, on the contrary, apparently does not begin to come into play vmtil six to eight weeks after antler growth has been initiated. This factor evidently exercises normally a role in the internal reconstruction and hardening of the July and August antlers by the laying down of the compact or cortical bone which re- places much of the initial spongy bone. Contrariwise, upon removal of the gonadal influence by castration, the capacity to form compact bone is abolished and shedding of the velvet never occurs. Whether the shedding of the velvet in the normal animal is directly con- trolled by testicular hormone is more difficult to answer. Olt"' regards the testicles as being directly responsible, basing his judgment upon the retention of the velvet after castration, as well as on the occurrence, during the normal shedding of the velvet, of a local "eosinophilia," the nature of which he does not further elucidate. Caton'" and Macewen,""' on the contrary, regard local George B. Wislocki 639 lactors as being responsible for the atrophy of the blood siipjjly to the antlers; these explanations assume in one way or another that the internal or external xessels are gradually mechanically choked by bone which is deposited around them. Macewen adopts the belief, current amongst naturalists and hunters, that the external blood supply of the velvet is obliterated by compression of the large vessels of the velvet at the base of the antler by the gradual enlarge- ment of the bony corona. Caton presents various reasons why these vessels are not compressed in that way, and offers, instead, the explanation that the in- ternal and external sets of vessels are directly obliterated by gradual closure of the smaller radicals in the Ha\ersian canals and marrow spaces, due to the pressure of accumulating deposits of lime salts. These theories in\olving local factors are based, however, more upon as- sumptions than upon any adequate histological proof. Vascular injections made by Rhtnnbler,"" as well as several which I have prepared," indicate that the blood supply reaches the antler almost exclusively by the external vessels in the velvet, whereas the internal stipply reaching the antler through its base is almost negligible. Thus the characteristic, dense calcification of the interior of the antler base is probably not responsible for the occlusion of any impor- tant larger blood vessels. Although Caton's and Macewen's concepts are poorly substantiated, it does not necessarily follow, as postulated by Olt, that atrophy of the velvet must be hormonally controlled; it is still conceivable that, in some as yet undisclosed way, local factors may be directly responsible for in- terruption of the circtdation. Nevertheless, the idea of hormonal control is inviting. In favor of it one might cite the slender parallels that the testes are known to exert effects upon certain vascular beds, for example, upon the vessels of the cock's comb (Hardesty""), as well as upon the human skin (Edwards and co-workers'"*). The eventual loosening and shedding of the dead antler, on the contrary, appear more definitely to be hormonally determined. At this period the pedicle becomes exceedingly hyperemic and the bone undergoes marked resorption, especially at the line of jtmction of the antler with the apex of the pedicle (von Korff,-" Macewen,"' Wislocki''). This event occius normally during the period of gonadal involution. Moreover, in deer castrated after the antlers are dead, the horns are shed prematurely within one to fotn- weeks following- removal of the gonads. Hence, the casting off of the antlers appears to be due to withdrawal of testictdar hormone, resulting in a local hyperemia and decalcification of the apices of the pedicles of the frontal bones. It has been pointed out that in all castrate deer, excepting the reindeer, the antlers assume variable forms, reaching their most extreme expression in roe deer in which bizarre masses of poorly calcified antler tissue arise. If castration occurs soon after antler growth has commenced, the horns develop normally up to the time when the influence of the testes ordinarily begins to express itself. Only from then on do the horns begin to show the effects of gonadal 640 Growth of Deer Antlers loss. At intervals in subsequent months and years additional abortive antler growth takes place from the antler pedicle. Apparently, since individual ant- lers can no longer be matured and shed, the nongonadal hormone which is responsible for initiating antler growth, and for stimulating the pedicle, succeeds only in producing an irregular succession of undersized, poorly calcified, abortive antlers which are not shed and which crowd one another. These become fused, forming eventually an irregular mass attached to an enlarged pedicle. Study of castrate antlers, besides observations on other types of deformed antlers (Schumacher*), indicates that the abnormal growth is initiated recurrently from the skin and periosteum covering the distal portion of the pedicle. Consequently the tissue covering the bony pedicle must be regarded as being peculiarly sensitive to some hormonal stimulus which is released periodically and which is not gonadal in origin. On the Factors Controlling the Periodicity of Antler Growth and Breeding Season. In several recent articles F. H. A. MarshalP''^''^ has assembled data to prove that the hypophysis is a regulator of gonadal rhythm. He concludes that, especially in female mammals, "the phases within the sexual cycle as well as sexual periodicity itself are brought in relation with the changes in the environment and more obviously with the seasonal environment by other factors, and there is now much general and experimental evidence to show that these factors are mainly or largely exteroceptive ones acting through the intermediation of the nervous system upon the anterior pituitary." He states further that in a large number of animals the incidence of daylight is almost certainly an important factor in controlling the cycle. He cites in this connec- tion data demonstrating the change over in the estrous cycle and breeding season in animals after transference across the equator. Especially illustrative of this are observations on ruminants, particularly sheep and deer. Regarding deer, numerous observations have been made by Donne^ upon various species of the northern hemisphere which have been transferred to New Zealand. The changes are well illustrated by the specific example of a group of red deer, as reported by Marshall,^" which were transferred from New Zealand to England: "Eight red deerhinds were brought from New Zealand where the deer rut in April to Warnham Court Park, Sussex, England, arriv- ing in March. They at first maintained their original gonadal rhythm and came on heat in April (at the New Zealand time). Their next sexual season, however, was in December-January and their third sexual season in England was in October or at the normal time for British deer. It is thus seen that the changed environment had an effect upon the periodicity of estrus, and in view of the evidence derived from other animals there is a presumption that the regulating influence was the anterior pituitary, this organ receiving stimuli from the environment and bringing the estrous cycle into suitable relation with seasonal change." George B. Wislocki 641 Marshall concludes, however, that ruminants, including deer, are excep- tional, since, if they react to light at all, it must be to diminution rather than to increase, because their breeding season, both north and south of the equator, falls in the period of diminishing daylight. If one be permitted to carry this interesting analysis and spectdation somewhat further, what apparently puz- zles Marshall about the ruminants can perhaps be harmonized to some extent with the behavior of other mammals. MarshalP"'^ does not consider the pe- riodicity of the antlers of deer, but it is clear from the observations of Donne^^ that, in the transference of deer across the equator, not only does the breeding season shift but the whole antler-gonadal cycle is also shifted to accord with the new climatic or seasonal environment. If we consider now the antler-gonadal seqtxence as related phenomena and as a whole, it appears jtistifiable to assume from the sequence observed in Virginia deer that the initiation of antler growth and the activation of the gonads occur in succession in the climatic spring or early summer. And if we assume on the basis of the accumiulated evidence that seasonal periodicity is related to exteroceptive stimuli, we may postulate with some validity from examination of various species of deer that the seasonal growth of the antlers and of the gonads is initiated sometime in the spring or early summer by the influence of light acting through the hypophysis.* Furthermore, it seems rea- sonable to assume froin the data on the growth of antlers and testes that the hypophysis first releases an antler-stimulating hormone which is followed one to two months later by the release of gonadotrophic hormones. The former stimulates periosteal hypertrophy in the antler pedicles and initiates the proc- ess of ossification. The latter activate the interstitial tissue of the testes and initiate spermatogenesis. The androgenic hormones, so stimulated in the testes, in turn subsequently influence and regulate the full maturation of the antlers. Finally, the casting off of the antlers occurs when the testes are declining, but is precipitated immediately if castration is resorted to in the fall; consequently dropping of the antlers must be due to a decline or removal of gonadal hormone. Regarding the relationship of initiation of antler growth to activation of the testes, it is apparent that in the European roe deerj- a similar situation prevails to that here observed in the Virginia deer. From perusal of data for this deer one finds that the antlers make their appearance from December to February (Brehm'°), whereas, according to Stieve,^" spermatogenesis does not commence until April. Thus it seems justifiable to conclude that, as in the * The pituitaries of the present series of deer have not been studied histologically. The adrenals, however, have been examined histologically by Dr. H. Stanley Bennett of this de- partment. He reports that no apparent seasonal variations are present in the specimens with regard to cortical zoning or staining of the cells, as revealed by Masson's technique. The thyroids of the same deer have also been the subject of study by Dr. A. L. Grafflin."^ uho re- ports that he has been unable to demonstrate seasonal variations in the material from that gland. f See footnote on page 635. 642 Growth of Deer Antlers Virginia deer, hypophysial stimulation of the antlers precedes gonadal stimu- lation by several months. Thus pituitary and gonadal hormones interplay in a definite way and se- quence in regulating antler growth. The gonadal hormone probably exercises a direct control in the internal reconstruction and maturation of the bony parts of the antler. As explained in a previous passage, it remains to be dem- onstrated whether it is also directly responsible for the vascular changes which lead to shedding of the velvet. The gonads appear also gradually to inhibit pituitary function, resulting in the eventual cessation of the hypophysial- gonadal cycle by mid- or late winter, with no further manifestation of hypophysial activity until the onset of the next cycle which begins, depending on the species of deer, several weeks or months after the previous antlers have been cast off. This inhibition of the pituitary is also illustrated by the fact that, upon castration of most species of deer after the velvet is shed, the antlers are cast within a month and new antlers are not formed until the following spring at the usual time (Caton'"). In similarly castrated roe deer, the antlers are likewise immediately shed, but, contrariwise, growth of the permanent castrate or wig antlers begins at once (Rorig^*), indicating that in this species the pituitary has not been fully inhibited. The tremendously rapid and ex- uberant growth of wig antlers in the roe deer, as compared with the lesser growth of castrate antlers in other deer, also suggests that the pituitary in this species is either relatively more potent in its antler-growth-promoting action or is less inhibited by the testes. In connection with the possibility of light stimulating the hypophysis of deer, the data of Schumacher™ are interesting. From the observation of roe deer, over many years in the Austrian Alps, he ascertained that their antlers were larger and better developed in those years in which the incidence of sunlight was greater. He concludes that sunlight should be regarded as the most important, although not the only, factor responsible for good develop- ment of the antlers. He believes that sunlight, activating vitamin D, is re- sponsible for the vigorous growth of the antlers in good years. He overlooks, however, the additional possibility of the direct response through neural cen- ters of the hypophysis to light. Other factors, including the abundance or scarcity of food, are also said to exert an influence on the size of the antlers (Caton,'' Seton,'^ Jacobi," Anthony''). Other aspects of the periodicity of antler growth and of the reproductive cycle fit into the concept of the role of the hypophysis and testes proposed above. There is a law governing antler growth, based upon the experience of numerous observers (Caton,'" Murie,"* Jacobi," Seton," Brehm,'" Rorig,'' '") to the effect that the more sexually mature and vigorous the animal, the earlier in a given season the antlers begin to grow (giaph 1). Thus the antlers of a six- or eight-year-old healthy buck put in their appearance some weeks earlier than those of the two- or three-year-olds. Similarly the velvet is shed sooner George B. Wislocki Qao and the antlers are cast earlier in mature bucks than in younger animals. It seems reasonable to interpret these observations in the sense that the more mature the animal the earlier in the season the hyjjophysis is activated and hence the sooner the hypophysis in turn activates the antler pedicles and the testes. In senescent animals (Seton/' Brehm/' Anthony''), conversely, the antlers diminish in size, appear later in the season, and the velvet is retained longer. In addition, in the senile roebuck the antlers are reported as never being shed (Schumacher"). These age changes indicate a lessening of the potency of the hypophysial-testicular mechanism. Finally, it is known that deer of a given species give birth to their young earlier in the southern than in the northern parts of the United States (Audubon and Bachman^'), and similar findings have been recorded for Europe (Brehm,^" Jacobi"), observa- tions which again point to seasonal influences in the control of reproduction. Moreover, many species of tropical and subtropical deer tend to have ex- tended sexual seasons and may breed at any time of the year (Marshall"'^, besides having no definite period for the appearance of the antlers (Brehm'"), observations which indicate that, in latitudes in which climatic changes are relatively slight, sexual periodicity and antler growth are not sharply regu- lated. Many of these various observations indicate the probable role of ex- teroceptive stimuli in regulating the periodicity of antlers and gonads and these leads should be more carefully followed up. We return at this point to the thesis of Marshall, to the effect that, whereas many mammals have their sexual seasons in spring or summer imder the exteroceptive influence of light, the ungulates have their period of rut or estrus in the fall, when, if exteroceptive stimuli are involved, the animals must be reacting to a diminution of light or possibly to temperature. From the observations reported in the present study, it appears very likely that in deer, as in many other seasonal animals, the hypophysial-gonadal cycle is also in- itiated by the exteroceptive stimulus of light. In ordinary seasonal mammals we are accumstomed to a short duration of these cycles, consisting of stimula- tion of the pittiitary gland followed immediately, in days or weeks, by the sexual season. In the white-tailed deer it is evident that the annual hypo- physial-gonadal cycle is initiated in early summer, the major difference be- tween deer and other mammals beino that in the former the entire cvcle requires five or six months for its completion, whereas in the latter the cycle is usually very much shorter. Thus it is apparent that in deer the pituitary- gonadal cycle is initiated, as in other seasonal mammals, in the spring or simimer, under conditions of maximal light, whereas only the terminal phe- nomena of rut occur in the fall at a time of diminishing light. That withdrawal of light may be an active factor in precipitating the terminal phase of the ungulate cycle is suggested by recent observations by Bissonnette'" to the effect that the onset of the sexual season in goats, which normally occurs in the fall, may be hastened by subjecting them to diminished light. If this be the case in 644 Growth of Deer Antlers ungulates in general, initial stimulation by light, followed by its subsequent withdrawal, may control the hypophysis and determine the sequence of the reproductive events. The present observations on deer draw attention merely to the fact that the annual hypophysial cycle has its onset much earlier than has been suspected from the sexual behavior and that, as in other mammals, light, acting in spring or summer, probably serves as the initial exteroceptive stimulus. It is suggested that similar conditions may prevail in other ungulates which exhibit sexual behavior in the fall. In conclusion, a few remarks seem appropriate regarding the question of secondary sex characters. We are accustomed to thinking of the anterior lobe of the hypophysis as acting upon the gonads, but as having no direct influence, in the adult animal at least, upon the Wolffian- or Miillerian-duct derivatives. These latter it is believed are activated by androgens and estrogens, and not directly by the anterior pituitary. Similarly it is generally assumed that the secondary sex characters are also under direct control of the gonads. Thus Riddle" states: "The secondary sex characters (in vertebrates at least) represent or reflect specific responses of developing parts, areas or tissues to a specific hormone of the gonad." Allen*" remarks: "Perhaps the time in evolution when the gonads assumed control of the secondary sex characters will never be known." Whereas Lillie" believes that "all sex primordia or rudiments and, possibly, all sex tendencies, assumed to be common to the life history of both male and female individuals of a species, may be divided in two classes, those that respond positively to the female hormone, and those responsive to the male hormone, each class being uninfluenced by the opposite hormone." Nevertheless, in the case of certain secondary sex characters, for example, the antlers, it seems likely that the initiation of their giowth in the adult male deer is under the control of the hypophysis. The subsequent maturation and involution of the antlers, however, appear to be subject to additional regula- tion exerted by the testes. Hence, it seems probable that this particular sec- ondary sex character is under the joint control of both anterior hypophysis and gonads in the manner outlined, with the pituitary providing the initial stimulus. In lower vertebrates most secondary expressions of sex appear to be under complete control of the gonads, yet occasionally in birds, as exem- plified in the plumage of one of the African weaver finches (Witschi"), the pituitary evidently plays a determining role. In mammals, in the case of the mammary glands (Turner^), the hypophysis may also exert a regulating in- fluence. One interesting difference exists between the case of the African weaver finch and the control of antler growth. Witschi's observations indi- cate that both gonads and plumage are controlled by gonadotrophic hormones. In the case of the Virginia deer, in which antler growth precedes gonadal growth by some six weeks to two months, it seems likely that two separate pituitary factors are operative, the first a form of growth hormone acting on George B. Wislocki 645 the skeletal tissues which give rise to the antlers, while the second consists of gonaclotrophic hormone activating the testes. The hypophysial and gonadal influences, here described as initiating and regulating antler growth, may not differ fundamentally from the hormonal control of growth of ordinary skeletal parts, for recent investigations demon- strate that androgens and estrogens, as the case may be, supplement hypo- physial factors in the normal process of ossification and matiuation of the growing skeleton (Silberberg and Silberberg^"-*'*^). Summary and Conclusions 1. The Virginia deer exhibits seasonal changes in the size and histological picture of the testes and seminal vesicles. Examination of the testes reveals the presence of spermatozoa beginning in July, being maximal in October, and diminishing in December and January. The seminal vesicles, inactive in June and July, commence to grow in August, and reach a high degree of ac- tivity by October. 2. The antlers begin their annual renew^al at a time when the testes and seminal vesicles are most inactive; they become hard and the velvet is shed when the testes and seminal vesicles are rapidly enlarging, while the antlers are lost when the reproductive organs have begun to decline. 3. Following castration, as reported in the literature, the antlers of most species of deer are retained permanently, continue to grow irregularly, and never lose the velvet. 4. Consideration of the normal seasonal cycle of antlers and gonads in the Virginia deer, and the restdts of castration of adtdt deer, indicate that the initiation of antler growth is independent of the testes. For various reasons presented in the text the assumption is made that the hypophysis, activated by light in the spring and early summer, is responsible for stimidating the growth of the antlers (May), and for the subseqtient activation of the testes (July). Subsequently the testicular hormone, either alone or more probably interacting with the hypophysial antler-growth-stimulating factor, induces the internal reconstruction of the antlers (August). The shedding of the velvet occurs after the antlers have hardened (September), but the evidence available at present is inconclusive as to whether the atrophy of the blood supply, re- sponsible for this event, is directly under hormonal control. Normal with- drawal of testicular hormone after the rutting season, or abnormally (following castration), produces hyperemia and decalcification of the pedicles, resulting in casting off of the antlers (January). The concept proposed here assumes that antler giowth is primarily initiated and controlled by the hypophysis, although the subsequent internal reconstruction and hardening of the antlers, as well as their ultimate shedding, are regulated by the testes, either alone oi in conjunction with the pituitary. 5. The gonads and male accessories, under the influence of the pituitary, 646 Growth of Deer Antlers begin to grow in the early summer during a period of maximal light, whereas growth is completed and rut ensues in the fall under conditions of diminish- ing light. It is pointed out that the annual hypophysial-gonadal cycle, com- pared with those of other mammals, is extremely protracted, and that rut represents a relatively brief terminal phase. It is postulated that, as in other seasonal mammals, the cycle is initiated by light, although the terminal phase of reproductive activity, including rut, may be influenced by withdrawal of light, as suggested by experiments on goats carried out by Bissonnette. 6. The growth of the antlers in adult male deer does not appear to be ini- tiated by the gonads, but by some other hormonal factor, probably hypo- physial. Hence, unlike many other secondary sex characters, antler growth is not primarily dependent upon the gonads. Only after the antlers have reached a certain size do the testes begin to exert a regulating influence. REFERENCES 1. Wislocki, G. B.: Anat. Recrd. 79(supp.):64, 1941. 2. Ziegler, L.: Benbaclituiigen i'lher die Brunst unci den Embryo des Relies (Hannover: 1843)- 3. Bischoff, T. L. W.: Entwickliingsgescliichle des Relies (Giessen: 1854). 4. Keibel, F.: Anat. Anzgr. i6(Ergzgshft.):64, 1899. 5. Keibel.F.: Anat. Anzgr. i9(Ergzgshft.):i84, 1901. 6. Retzius, G.: Anat. Anzgr. i6(Ergzgshft.):65, 1899. 7. Retzius, G.: Biologische Untersuchungen (n.F.) IX (Stockholm, Jena: 1900), 109. 8. Strahl, H.: Anat. Anzgr. i6(Ergzgshft.):66, 1899. 9. Lenninger, W.: Ztschr. f. Anat. 68:230, 1923. 10. Stieve, H.: Ztschr. f. mikroskop.-anat. Forsch. 13:159, 1928. 1 1. Wislocki, G. B.: Amer. Jl. Anat. 71:371, 1942. 12. Caton, }. D.: The Antelope and Deer of North America (New York: 1877). 13. Seton, E. T.: Life-Histories of Nortliern Animals, I (New York City: 1909). 14. Townsend, M. T., and Smith, M. W.: Roosevelt Wild Life Bull. 6: i6i, 1933. 15. Gaskoin, J. S.: Proc. Zool. Soc. London 24:151, 1856. 16. Fowler, G. H.: Proc. Zool. Soc. London 1894:485. 17. Rorig, A.: .A.rch. f. Entwcklgsmech. d. Organsmn. 8:382, 1899. 18. Rorig, A.: Arch. f. Entwcklgsmech. d. Organsmn. 23:1, 1907. 19. Brehms Tierleben, 4th ed., Saugetiere, IV (Leipzig und Wien: 1916), 80. 20. Tandler, J., and Grosz, S.: Die Biologischen Grundlagen der Sehnndliren Gescldechl- scharaktere (Berlin: 1913). 21. Tandler, J.: Anzgr. Kais. Akad. d. Wissnschftn. (Wien), malhem.-naturwissnschft. Kls. 47:252, 1910. 22. Jacobi, A.: Zool. Anzgr. 96 (Ergzgsbd.), 1931. 23.01t, A.: Brcht. Oberhess. Gschft. f. Natur- u. Heilk. zu Giessen (n.F.: naturwissnschft. Abtl.) 11:3, 1927. 24. Olt, A.: Natur u. Mus. (Brcht. Senckenberg. naturforsch. Gschft.) 58:376, 1928. 25. Macewen, W.: The Growtli and Shedding of the Antler of the Deer (Glasgow: 1920). 26. Rhumbler, L.: Wilhelm Rous' Arch. f. Entwcklgsmech. d. Organsmn. 1 19:441. 1929. 27. Hardest)', M.: Amer. Jl. Anat. 47:277, 1931. 28. Edwards, E. A.; Hamilton, J. B.; Duntley, S. Q., and Hubert, G.: Endocrinology 28:119. 1941- Geoi2re B. Wi slock i & 647 29. von Rorff. K.: Anat. Hfte. 51 :6gi, 1914. 30. Schumacher, S.: Osteneichs Weidwerk 2:1, 1937. 31. Marshall, F. H. A.: Philos. Trns. Roy. Soc. London, s.B 22():423, 1936. 32. Marshall, F. H. A.: Proc. Roy. Soc. London, s.B 122:413, 1937. 33. Marshall, F. H. A.: in Les Hormones Sexuelles, ed. by L. Rrouha (Paris: 1938), 24-. 34. Donne, T. E.: Tlie Game Animals of Nciu Zealand (London: 1924). 35. Grafflin, A. L.: Jl. Morphol. 70:21, 1942. 36. Schumacher, S.: Deutsch. Jiiger 45: 1, 1937. 37. Anthony, H. E.: Bull. New York Zool. Soc. 32:3, 1929. 38. Murie, O. J.: North Amer. Faun. no. 54, 1935. 39. Audubon, J. J., and Bachman, J.: The Q_uadrupeds of North America, II (New York: 1851) 220; III (New York: 1854), 168. 40. Bissonnette, T. H.: Physiol. Zoo). 14:379, 1941. 41. Riddle, O.: in Sex and Internal Secretions, ed. by E. Allen (1st ed.; Baltimore: 1932), 263. 42. Allen, E.: in Sex and Internal Secretions, ed. by E. Allen (1st ed.; Baltimore: 1932), 460. 43.Lillie. F. R.: in Sex and Internal Secretions, ed. by E. Allen (1st ed.; Baltimore: 1932), 9. 44. Witschi, E.: Scientia 60:265, 1936. 45. Turner, C. \V.: in Sex and Internal Secretions, ed. by E. Allen (2d ed.; Baltimore: 1939), 740. 46. Silberberg, M., and Silberberg. R.: Archs. Pathol. 32:83, 1941. 47. Silberberg, M., and Silberberg, R.: Anat. Recrd. 80:347, 1941. 48. Silberberg, M., and Silberberg, R.: Amer. Jl. Anat. 69:295, 1941. EXPLANATION OF PLATES PLATE 1 Fig. 1. A section of the testes of a June deer, x no. Fig. 2. A section of the epididymis of the same deer, showing the absence of sperm in the tubules, x i lo. Fig. 3. A section of the testes of an October deer, x 165. Fig. 4. A section of the epididymis of the same deer, showing abundant sperm in the tubules, x no. [648] r ^- *ji«i • '''-'-•i^^^*^-."^;-^;^^- mSS Plate 2 PLATE 3 Fig. 9. The seminal vesicles of an October deer, x 90. Fig. lo. The same enlarged, x 250. [652] Plate 3 SEX DIFFERENTIATION IN HETEROGENOUS PARABIOTIC TWINS (A?7ibystomax Triturus) By EMIL WITSCHI AM) HARRIET M. M. McCURDY From the DEPARTMENT OF Z05lOGY, STATE UNIVERSITY OF IOWA IOWA CITY, IOWA SEX DIFFERENTIATION IN HETEROGENOUS PARABIOTIC TWINS (AmbystomaxTnturus)* WITH growing information on sex development in parabiotic twins, definite indications of taxonomic specificity of the sex-differentiating substances are increasingly demanding our interest. It has been found (Witschi^'*) that in some species these substances, corticm and medullarin, are spread by the blood stream like hormones, wdiile in others they diffuse more slowly and with falling concentrations through the tissues. In at least one amphibian species, their presence was noticeable even before the initiation of morphologi- cal sex differentiation (Witschi and McCurdy^). On the contrary, in all other well-analyzed cases, the antagonism between the sexes appears only at later developmental stages (Witschi,^* ^comprehensive review®). This variety of conditions immediately suggests further investigations by the method of heterogenous parabiosis. It would seem interesting indeed to know what effects might be obtained from inductive substances of sala- manders, poured into the blood streams of frogs. Unfortunately, the possibili- ties of such testing are limited by the factors of compatibility in heteroplastic grafting. In our laboratory, salamander-frog and newt-frog twins lived for several weeks, but they died invariably before the stage of initial sex differen- tiation. More successful were the experiments with salamander-newt twins. The fact that grafting between taxonomic orders (Caudata and Salieritia) and families (Salamandridae and Anihystoynidae) yields twin combinations which may last for weeks and even years is in itself remarkable, and deserves further study; the present paper will, however, be restricted to the analysis of the interactions observed in the sex development of the heterogenous twins. Controls In these experiments, three species of salamander, Ambystoma maculatiim, A. jeffersonianum^ and A. tigrmum, were used in combination with the Cali- fornian newt, Triturus torosiis. The latter is smaller than Ambystoma tigri- num, but slightly larger than A. m,aculatiim. This is true particularly for the late larval and metamorphosing stages, while adult newts and spotted sala- manders are nearly of same size. A. jeffersonianiwi in Iowa is represented by a series of size variations. A dwarf form from Cedar Falls has been described in 1937.^ The typical form is larger than T. torosiis and A. macxilatum, but * The first cxpeiiments ol this series were started in 1927 while, as a fellow of the Rocke- feller Foundation, the senior author spent three months in the Anatomy Department of the University of California. In spite of the length of the interval, the aiUhors are desirous of expressing their gratitude for the cordial hospitality extended to them hy Dr. Herbert M. Evans and for the stimulating interest and the generous help with \vliich he furthered their work. During several years the continuation of the experiments was made possible by grants from the National Research Council, Cominittee for Research in Problems of Sex. C657] 658 Sex in Parabiotic Twins not quite of the size of A. tigriniim. It appears only once as a mate of a parabiotic pair (jef-T^; fig. 7). All other jeffersonian salamanders of the series (table 3) belong to a medium-sized variety, the eggs of which were collected in the vicinity of Iowa City. Few observations on sex development in jej- TABLE 1 Parabiotic Twins, Series mac-T {Ambystoma maculatumxTriturus torosus). F, normal ovaries; f, inhibited ovaries; M, normal testes; m, inhibited testes; i, indifferent gonads. The maculatum members are of the semi-differentiated Ozark race except those of pairs 1 2 and 16, which are from Massachusetts and Tennessee. No. I 1 3 4 5 6 7 8 9 II 12 14 15 16 17 19 20 21 22 Sex glands M M F M F F M F F M f-F F M F F F M Trilurus i(f) m M m f M i(f) M M m f M m f M M M f m Age, mo. 5 5 4 5 5 9 8 5 4 5 8 7 7 7 8 5 5 5 8 Size, mm. A. mac. 34 47 42 40 36 46 40 41 39 23 41 41 34 3 16 37 39 40 Triturus 32 31 ^2 23 26 32 29 29 26 35 53 26 36 29 50 43 21 21 36 jersoy-iiajium controls are available. They indicate a slight hermaphrodite tendency, at least in some of the varieties. Single newts and salamanders metamorphose at the age of from four to six months, and sex differentiation starts during the second month, when the larvae are from 22 to 30 mm. long. Tiger salamanders progress somewhat faster than /I. maculatum, A. je^ersonianum, and T. torosus, though time differences are not sufficiently large to enter as an important factor into the consideration of the experiments reported in this paper. At the metamorphosis stage, the gonads are largest in A. tigmium and jeQersonianiim , smallest in T. torosus (compare the cross sections shown in fig. 8m, ^T, and 12/). Triturus torosus differs from the salamanders in being strictly gonochoristic. No trace of hermaphrodism has yet been found in any controls of this species (McCurdy"). On the other hand, in the salamanders, \ Witschi and McCurdy 659 hermaphrodite features are observed with striking regularity (Gilbert""), most frequently in the Ozark race of A. maculaturn, which is represented in seven- teen of the nineteen maciilatum-Triturus pairs (table i). Besides the single normals, the homogenous parabiotic pairs of the various species must also be considered as controls of the experimental heterogenous twins. In Triturus torosus, the sex glands of female-female and male-male twins develop normally, like those of single controls. In female-male com- binations, one observes a severe antagonism from which, at the beginning, both sexes suffer almost equally. It becomes evident even before morphologi- cal sex differentiation, and interferes with the formation of the indifferent gonad primordia. Later, the male recovers normalcy while the female be- comes nearly or completely sterile (Witschi and McCurdy''; detailed paper in preparation). In heterosexual pairs of salamanders, the antagonism is less pronounced, and only ovarial development suffers appreciably. The Ozark race of ^. maculatiim is especially notable for a very low degree of interference between members of heterosexual pairs (Witschi,* " the latter contains refer- ences to the important papers of Burns and Humphrey). Experimental Series The Californian newt, Triturus torosus (family Salamandridae), was joined with each of the three mentioned species of the genus Ambystoma (family Ambystojnidae). The three combinations will be designated by the symbols mac-T, tig-T and jef-T. Embryos of the tailbud stage were grafted together by the previously described routine technique (fig. 4). In early experiments, one embryo was joined with its right side to the left of another (figs. 5 and 7); later it was found to be of some advantage to use the right side in both em- bryos for grafting (figs. 4 and 6). At first, no unusual difficulties appear to exist. Healing and recovery from the operation are as easy as in homogenous twins. Later on, however, the salamander remains immotile for two or three weeks, due to a poisonous stibstance contained in the newt (Twitty"). This paralysis of the salamander during the critical period when feeding habits normally become established, is probably the main reason for the excessive mortality in this experimental group. Of about eight hundred pairs of embryos which were successfully joined, only thirty-nine survived until the time of metamor- phosis or longer. Anastomoses between large blood vessels become established very early (fig. 5), and even at the metamorphosis stage, blood is still freely exchanged. The largest pair was preserved at an age of thirteen and one-half months; it had passed through metamorphosis when six months old. Since the change from the acjuatic to the terrestrial life proved to be a great hazard, most pairs were preserved toward th.e end of the metamorphic period. As a rule, the newt is the one of the pair which accepts food first. Later, after recovery from the effect of the poison, the salamander also begins to eat and sometimes even assumes the leading role both in swimming and eating. Like normal sala- manders and newts, most twin pairs metamorphose at an age of from four to 66o Sex in Parabiotic Twins six months (room temperature); though poor feeders may be delayed, some as much as two or three months. In only one of the thirty-nine pairs were both mates of approximately nor- mal size at the time of metamorphosis (jef-T^; fig. 7). Quite obviously, there exists an imbalance in these twin systems, tending toward dwarfing or even elimination of one of the partners. Surprisingly enough, Triturus is the loser in thisstruggle when combined with ^. maculatum (fig. 1), but the winner here in competition with the normally larger A. jeffersonianum and A. tigrinum (figs. 2, 3). In considering the measurements given in tables 1-3, one must take into account the fact that the normal size of tiger salamanders at the meta- J Fig. 1. Heterogenous parabiotic pair, Ambystoma maculatum x Triturus torosus, at the end of the metamorphosis stage. The salamander (white) is larger and the newt (stippled) smaller than normal size. Four-sevenths actual life-size. Fig. 2. Heterogenous pair jef-Tj^, at a late metamorphosis stage. The salamander is of pigmy size, but its testes are of about normal dimensions (the folded organ shown at the right side). The newt is rather larger than normal, but its gonads are reduced to narrow vestiges (center figure). Twins are four-sevenths life-size, sex glands x 6. Fig. 3. Heterogenous pair tig-T,, at a late metamorphosis stage. The dwarfed salamander (stippled) has ovaries of normal size and structure (left figure). The newt has undifferentiated gonads of rudimentary size (center figure). Twins are four-sevenths life-size, sex glands x 6. morphosis stage is from about 140 to 180 mm. Even the case tig-Tg (table 3) is, therefore, no exception from our generalization, for the newt of 50 mm. is above the average size of controls, while the salamander of 90 mm. is under- sized. The reasons for this peculiar growth phenomenon are not clear, but it is significant that the developmental rates of the sex glands do not follow the same trends as the soma. The excessive mortality rate seems to influence the initial sex ratio but very slightly. Adding up all surviving salamanders, we find twenty-two females and fourteen males. The moderate prevalence of females is in line with pre- vious similar observations on somewhat greater vitality of this sex in parabiotic Witschi and McCurdy 661 TABLE 2 Parabiotic Twins, Series jef-T {Ambystoma jeffersonianum-xTriturus torosus). F, normal ovaries; f, inhibited ovaries; M, normal testes; m, inhibited testes; i, indifferent gonads. The A.jefersonianiim was of the typical Iowa form in pair i and of a slightly dwarfed varietv in all other pairs. No. I 4 5 6 7 8 9 10 II 12 13 Sex glands A. jej. F F M F F M M F M F Triturus M f i(m) m i(m) i(0 i(0 i(m) F i(0 i(0 Age, mo. 13 9 9 7 10 4 5 3 4 5 5 11ZC, mm. ^. ;,./. 98 15 12 19 I I 24 21 17 6.5 18 15 Tritur 41 45 48 40 32 44 40 35 45 55 <;! TABLE 3 Parabiotic Twins, Series tig-T {Amby stoma tigrinitmxTriturus torosus). F, normal ovaries; f, inhibited ovaries; M, normal testes; m, inhibited testes; i, indifferent gonads. The A. tigrinnni member of pairs 1-8 is of the mexicanum variety, the one of pair 9 is of the typical Iowa form. No. I 2 3 4 5 6 7 8 9 Sex glands //. in F F M F M F F F M Triturus i(0 i(m) i(m) i i(0 i(m) i(m) ilO i(0 Age, mo. 7 7 9 8 12 9 16 16 6 Size, mm. 36 -5 8 21 24 19 90 Triturus 19 49 16 53 22 50 50 50 pairs (WitschP). The sex of Triturus cannot always be ascertained, though the definite identification of eight females and fifteen males proves that also in the newt there occurs no pronounced selective elimination. If, on the basis of the observations on controls, one might be tempted to predict a leading role for the newt members, the actual results prove mark- edly different from such expectations. As a matter of fact, a survey of the entire material which is contained in condensed form in tables 1 to 3 shows 662 Sex in Parabiotic Twins that neither ovaries nor testes of the salamanders become much modified, while the gonads of the newts suffer severely in all but one of the twin com- binations. TJ}e maculatum-Triturus series (table i) is of particular interest becavise here the antagonism is moderate enough to permit identification of the ge- netic sex of the newt in every case. In combinations of female salamanders with male newts (mac J-T ^), both twins have essentially normal glands (fig. 9). The newt testes are not smaller than in single controls of corresponding develop- mental age. The salamander ovaries are entirely normal in five of the six cases of this group (see table 1). They contain numerous large ovocytes and spacious ovarial sacs. In the sixth pair (mac-T,.), the left ovary is likewise normal, while the right one has no open ovarial sacs. Instead, its medullary cords are compact (fig. lomj, and in a few places, they contain even a few primary gonia (fig. 10/77^,). However, at the same time, this ovary contains large and healthy ovocytes, so that one hesitates to ascribe the deviation to the proximity of the newt testes. Yet it may be significant that we deal here with the oldest pair of this series (table 1). A delayed suppressive action of the Triturus testes is, therefore, not excluded. In fact, it gains support from another case, the first of the jeffersonianum-Triturus twins (jef-T^). This differs markedly from the other pairs of the second series (table 2) but resembles closely the mac-T pairs. Probably the salamander in this instance belongs to a racial strain of marked hermaphrodite tendency, while all others are of more gono- choristic type. At any rate, in this jef-T^ pair we find also newt testes of normal size and salamander ovaries with some effects of inhibition (fig. 11). As in mac-T,., the ovary which is located nearer the newt testes shows more degen- erative changes in the cortex and has more medullary gonia than the other one. These two cases, each the oldest pair in its series, present the only, and certainly very meager, indication of a possible inductive influence coming from the newt and affecting the salamander sex glands. In female-female pairs macJ-T 2)> the ovaries of the newts are small, re- tarded, and of reduced fertility. In male-male twins (mac J^-T J*), the Triturus testes are considerably inhibited and, therefore, much smaller than in the mac 5-T ^ combination. Lastly, in the mac J'-T 5 combinations, the sala- mander males completely suppress ovarial development in newt females (fig. 8). The gonads of the latter are nearly sterile primitive folds in which the female character is barely recognizable. In short, of the reciprocal hetero- sexual combinations, one (macJ-TJ^) shows the minimal, the other (macj^- TJ) the maximal degree of interference and inhibition. In the unisexual combinations, the gonads of the Triturus member are partially inhibited. In two instances (nos. 17 and 19), the maculatum members, which had been partly resorbed, were without gonads; their Triturus co-twins were males with normal testes. The jeffersonianum-Triturus and tigrininn -Triturus series are summarized in tables 2 and 3. The sex glands of the newt are much more severely damaged Witschi and McCurdy 663 in the combinations with jeffersoniaruim or iigriniirn salamanders. As a rule, they consist only of long, nanow, and threadlike folds (figs. 2 and '^) which, on sectioning, prove to contain but few primordial gonia (fig. 12T) and to be sexually undifferentiated. In the tables, the symbol i (indifferent) is followed by / or 7)1 if a majority of the germ cells are found in either the cortical (as in fig. ST), or the medullary (fig. 12T) position. This relationship is not suf- ficient, however, to establish with certainty the basic sex; for in normal newts, at the indifferent stage, the gonia are also scattered throughout the gonad primordia. In jef-T^, the newt glands are definitely on the way to ovarial differentia- tion, a few germ cells having become transformed into synaptic ovocytes. Similarly, in jef-T,., a nimiber of gonia are clearly located in medullary cords and some segments of the glands have a distinctly testicidar appearance. In one pair, the newt gonads are nearly normally developed. This is the pre- viously described case of jef-T^ (figs. 7, 11), in which the salamander has fairly large ovaries and the newt good-sized testes. These three cases form an im- portant link between the first and the two other series, for in the seven re- maining jef-T and in all tig-T pairs the newt gonads are reduced to diminutive primordial rudiments, while the salamander gonads are of the impressive size (figs. 2, 3 and 12), which is characteristic for normal specimens of these two species. Discussion The present series of heterogenous twins differs markedly from the many previously described groups of homogenous twins or combinations of various races or species of salamander, especially in the following two features. 1. Tritiirus does not interfere with normal gonad development of Am- bystoina co-twins, in any of the four possible sex combinations. 2. Amhystomas of the three species inhibit the development of co-twin gonads not only if the newt is of opposite sex, but also if it is of same sex. The first result is very smprising in view of earlier work which had given proof that the newt is a distinctly gonochoristic species and that its gonads emit morphogenic substances of the corticin and medullarin type (Witschi and McCurdy''). As stated before, in heterosexual pairs of newts (TJ-TJ*), the severe mutual antagonism becomes manifest even before morphological sex differentiation, that is, at so early a stage, that the failure of the same species to produce effects when in parabiosis with salamanders cannot be at- tributed to a time factor. The lack of effectiveness of the newt substances is drastically demonstrated by the five tnaculatum^-Trituriis^ pairs (fig. 9), in which large testes of Tri turns coexist with large and absolutely normal Ambystoma ovaries. In Triturus^-Tritiirusi^ pairs, testes of ecjual size and normalcy reduce the female gonads to near sterile, and often entirely undif- ferentiated, rudiments. It must, therefore, be concluded that the inductive sex substances of Tritiirus torosus are of a specific nature which prevents them from becoming effective in the salamander co-twins. Since blood is exchanged 664 Sex in Parabiotic Twins through large anastomoses (fig. 5), it seems improbable that the substances become removed before they reach the gonads of the co-twin. It appears more likely that the salamander gonad is not an adequate receptor and therefore remains unresponsive to the Triturus substances. The fact that in two of the oldest pairs (mac-Tg and jef-T.^) the ovaries of the salamander show some deficiencies may possibly indicate a change of con- ditions governing sex interference on approach of the stage of maturity. How- ever, the observed effects are too slight and the material too meager to permit a definite statement in this respect. Attempts to solve this particular problem by transplanting gonads between larvae of salamanders and newts were un- successful, because the implants failed to develop properly. The second result is more complex in its aspects and implications. The fact that the newt gonads in most jeffersonianum and in all tigriniim combina- tions, even the unisexual ones, are extremely reduced could seem to indicate that one deals with a general effect of parabiosis between species of different taxonomic families, rather than with specific sex antagonism. It is easy to show that such an assumption would be untenable. For one thing, in the maculatum combinations, where the antagonism is more moderate, we have a gradation of inhibition types, which definitely depends on sex combinations. Inhibition becomes more and more severe in the following sequence: (a) mac5-T ^, (b) mac$-T $, (c) macJ'-T (^, (d) mac^-T 5. Secondly, inhibition effects happen to be more pronounced in newts combined with salamanders of large gonad and small body size (Jeff, and tig.) than in those joined with salamanders of large bodies and relatively small gonads (mac). Hence, the extent of the reaction in the newts is not correlated with body size of the salamanders. Thirdly, in three pairs in which the salamander member was without gonads (due to early destruction of the gonad-forming body region), the newts had normal and full-sized gonads. The pair jef-T^^ is the most con- vincing case in this regard. Its Triturus member is the only female newt with entirely normal ovaries in the whole set of thirty-nine twins. It is significant that the jeffersonianum member is deficient mainly in the lower body region while its upper parts, the head, and especially also the hypophysis gland, are as well developed as in most other jef-T pairs. The importance of this case gains relief by comparison with pairs like jef-T. Avhere the upper body region of the salamander is deficient, the hypophysis absent, and the testes well de- veloped; in this pair the Triturus gonads are as rudimentary as in all other combinations with male salamanders. The conclusion that the salamander gonads alone are responsible for the inhibitions is unavoidable. The interest- ing suggestion made by C. R. Moore,^° in connection with his work on the opossum, namely that substances issuing from the entire somatic complex rather than from the gonads only may be the humoral agents responsible for free-martin effects, is therefore ruled out as far as the present experiments and materials are concerned. Likewise, the mentioned observations show that the hypophysis gland is not in any specific way involved. In fact, some other Witschi and McCurdy 665 unpublished experiments prove that even loss of tlie hypophyses by both twins does not change the larval sex antagonism. Having therefore made sure that we deal actually with gonadal antago- nism, it becomes immediately evident that not only do the sex glands of Triturus respond to the gonadal substances of the Ambyslomas, but that thcv serve as extremely sensitive indicators. This is in keeping with the previously established rule that heterosexual parabiosis produces more extensive reac- tions in the gonochoristic species and races than in those with hermaphrodite tendencies (Witschi*). Sensitivity grows from A. maculatum to A. jefferson- ianum, A. tigrinum, and T. torosus. It fits well with this rule that maculatum males and females reduce the gonads of Triturus more than those of twins of their own kind; or that the female of A. tigrinum, which manifests a domi- nance over male salamanders only under special conditions of time and size (Witschi"), is able to completely suppress testicular differentiation in Triturus males. Yet there still remains the problem of unisexual antagonism which is ap- parent in the suppression of Triturus ovaries by Ambystoma ovaries and of Triturus testes by Ambystoma testes.* Of course, it is quite probable that every larval gonad releases simultaneously corticin as well as medullarin, albeit in different proportions. At least for all forms which, like the sala- manders, are strongly inclined toward hermaphrodism, it is safe to make this surmise. All differentiating gonads consist morphologically of double pri- mordia: cortex and medulla. Sex differentiation consists, in morphological terms, in the further development of one and the relative reduction and sup- pression of the other. It can, therefore, be taken for granted that in female pairs a small amount of medullarin, released by the Ambystoma ovary, is retarding ovary formation in Triturus, and likewise in male pairs, that corticin given off by the salamander testes holds back the development of male gonads in a newt co-twin. In view of the previously established high responsiveness of Triturus gonads, this new aspect of the ambisexual nature of vertebrate sex glands certainly must be given consideration as a factor in the interpreta- tion of unisexual gonadal antagonism. Yet it cannot in itself serve as a com- plete and satisfactory explanation, since there still remains the odd fact that in the maculatum-Triturus combinations, the Triturus testes are more in- hibited by salamander testes than by salamander ovaries (table i). It certainly seems unlikely that male salamanders produce more corticin than females of the same species. Therefore, as in an earlier discussion of conditions ob- served in heterogenous frog twins (Witschi"), w'e feel again impelled to assume that the growth of certain organs, and especially the sex glands, is controlled by special growth substances which are available only in limited quantity. They are possibly responsible for compensatory hypertrophy after partial castration, as well as for the often-experienced difficulty of establishing suc- * According to a recent paper by Humphre)" similar conditions occur in A. maculatum hosts carrying large gonads derived from implanted A. ligriyium preprimordia. 666 Sex in Parabiotic Twins cessful grafts of gonads in normal animals as compared with the relative ease of getting implants in castrates. In many heterogenous parabiotic twins, the ovaries (or testes) of one twin seem to be able to monopolize the whole supply, thus depriving the co-twin more or less completely of this substance and, TABLE 4 Type of Reactions Produced in the Sex Glands of Male and Female Newts {Triturus torosus) by Parabiotic Twins F, M, normal sex glands (no reaction); f, m, inhibited yet clearly differentiated sex glands (moderate reaction); i, rudimentary and sexually undifferentiated glands (maximal reaction). The sex types of the producers of inductive substances are indicated by ss (distinct hermaph- rodite tendency), s (slight hermaphrodite tendency), d (prevailingly gonochoristic) and dd (fully gonochoristic). Producers of inductive agents Reactions observed in Species Sex type Triturus cf Triturus 9 A. maculatum 9 A. jeffersonianum 9 A. jefersonianum, and A. tigrinum 9 ss S d M m i f f i A. maculatum cf SS s,d m i i A. jeffersonianum, and A. tigrinum (^ i Triturus torosus 9 dd dd m M M F Triturus torosus cf i therefore, of the possibility of growing normal gonads. These hypothetical substances are not hypophysial hormones, since sex glands of larval newts grow normally even in the complete absence of the hypophysis gland (Woron- zowa and Blacher,^" Witschi''). Nor can they be sex hormones, for the charac- teristic reactions which these produce in salamanders (Foote") are entirely different from anything observed in the present experimental series (Witschi'"). Since they serve to control the size and the total amount of gonadal tissue in single individuals, as well as in conjoined twins, they may be designated, provisionally, as ovarial and testicular groxoth substances. The reactions produced in Triturus testes and ovaries are represented in a summary way in table 4. Their correlation with sex (5, (^) and sex type (ss, s, d, dd) of the parabiotic twins is quite obvious. Comparing the reactions in heterogenous pairs with those obtained in homogenous Triturus twins (two bottom columns), it becomes evident that in the latter the effects are produced only by a mutual cortex-medulla antagonism; whereas in the former the mechanism is complicated by the following factors: 1. Salamanders do not respond to inductive agents of the newt and, there- fore, continue with the unimpeded output of substances which interfere with gonadal development in the newt. Witschi and McCurdy 667 2. Both sexes of the salamander produce corticin as well as niedullarin— though at different ratios— and suppress thereby ovarial as well as testicular development of newt co-twins. 3. In order to understand the reactions in unisexual pairs, one is forced to assume that gonad size is limited by special ovarial and testicular growth stibstances. These substances seem to be utilized and used up more quickly by tlie salamanders than by the newts. The unequal growth of the gonads in heterogenous twins is not a unic^ue phenomenon. In the above-mentioned frog twins (R. sylvalica x R. tempor- aria), it was found that the spleen of the sylvatica member is always greatly enlarged, while in the temporaria it is reduced to a mere vestige. In the present experimental series, we find that in general body growth, Ambystoma macu- latum surpasses Triturus, while the other two salamander species lag behind their newt co-twins, especially during the late larval stages. Of great interest is the development of the kidneys. In the mac-T combinations, the kidneys are somewhat larger and better differentiated in the maciilatum than the Triturus member. In all jef-T and tig-T pairs, however, the Triturus kidneys alone are normally developed, while the salamander kidneys are inhibited, of small size and histologically incompletely differentiated. There is no cor- relation between size of the kidney and size of the sex glands. Kidney develop- ment and body size run somewhat parallel, though in jef-T^, the salamander is of normal size while its kidneys are greatly inhibited. Summary 1. Of about eight hundred pairs of salamander-newt twins which were tuiited in parabiosis at the tailbud stage, only tliirty-nine lived to tlie metamorpliosis or later stages. 2. In heterogenous pairs, growth of the twins is usually uneven: so that one member is of normal size, or even larger, while the other member is dwarfed. 3. Similarly, the kidneys of the two species develop differently. 4. It is stiggested that the growth of various organs is controlled by specific growth substances, which are produced in limited amotmts. In heterogenous parabiosis, one twin may use up more than its share of such substances. 5. With respect to sex, the three species of salamanders used in this experi- ment are of markedly different type. Ambystoma maciilatum shows distinct hermaphrodite tendencies (ss). A. jeffersonianiim is of slightly hermaphrodite (s) or prevailingly gonochoristic (d) type. A. tigrinum is prevailingly gono- choristic (d). The newt. Triturus torosus, with which these salamanders Avere combined, is of distinctly gonochoristic type (del). 6. Sex development in salamanders is not modified by parabiosis with newts. It is concluded that the sex-differentiating substances, corticin and medul- larin, have a certain taxonomic specificity and that salamander gonads do not respond to the substances emitted by the newts. 668 Sex in Parabiotic Twins 7. Sex development in the newt is greatly affected by parabiosis with sala- manders. The detailed analysis leads to the conclusions: a. That both sexes of salamanders release corticin as well as medullarin, though in different proportions and b. That the growth of ovaries and testes is controlled by gonadal growth substances. REFERENCES i.Witschi, E.: Jl. Expei . Zool. -,8:113, 1931. 2. VVitschi, E.: in Sex and Inlcrnal Secretions, ed. by E. Allen (ist ed.; Baltimore: 1932). 160. 3. Witschi, E., and McCurdy, H. \l.: Pioc. Soc. Exper. Biol, fl- Med. 26:655, 1929. 4. Witschi, E.: .'^nat. Recrd. 66:483, 1936. 5. Witschi, E.: Jl. Exper. Zool. 75:313, 1937- 6. Witschi, E.: in Sex and Internal Secretions, ed. by E. Allen (2d ed.; Baltimore: 1939), 145. 7. McCurdy, H. M. M.: Amer. Jl. Anat. 47:367, 1931. 8. Gilbert, W. M.: Proc. Iowa Acad. Scis. 47:307, 1940. 9. Twitty, V. C: Jl. Exper. Zool. 76:67, 1937. 10. Moore, C. R.: Physiol. Zool. 14:1, 1941. 1 1. Humphrey, R. R.: Amer. Jl. Anat. 70:345, 1942. 12. Woronzowa, M. A., and Blacher, L. J.: Wilhelm Roux' Arch. f. Entwcklgsmech. d. Organsnin. 121:327, 1930. 13. Witschi, E.: Proc. Soc. Exper. Biol. & Med. 27:763, 1930. 14. Foote, C. L.: Jl. Exper. Zool. 86:291, 1941. 15. Witschi, E.: Cold Spring Harbor Sympsa. Quantit. Biol. 10:145, ^9A-- \ EXPLANATION OF PLATES PLATE 1 Figs. 4-7. Heterogenous parabiotic pairs. ¥ig. 4^. Ambysto)7ia niaculaliun x white axolotl, one day after grafting opera- tion; showing the stage at which union is established, x 8. Fig. 5. Two pairs {Ambystoma maculatum X Triturus torosus) showing the reduction of the yolk sacs and formation of anastomoses between the abdominal veins. The (longer) salamanders are paralyzed. X 5. Fig. 6. Another pair of the same combination at the time when feeding be- comes established. The salamander (longer member) is coming out of the paralyzed condition, x 4. "]. Ambystoma jefjersonianum X Triturus torosus, shortly after metamorphosis (6 months old), x 1. [670: ■J V w Plate 1 C671] PLATE 2 Figs. 8-1 o. Sex glands of three pairs of the combination Amhystoma maculatum X Triturus torosus. m, gonads of A. maculatum; T, Triturus gonads. All x 188. Fig. 8. Case mac-Ti: normal testis of salamander, rudimentary gonad of newt. Fig. 9. Case mac-Tgi normal ovary of salamander, normal testis of newt. Fig. 10. Case mac-T,;: Slightly reduced ovary of salamander, normal testis of newt. 1:672] Plate 2 PLATE 3 Fig. 11. Ambystoma jeffersonianum $ x Triturus torosus (^ (jef-Ti.); cross section through gonads at the age of 13 months, x 94. Fig. 12. Ambystoma tigyinimi (^ x Triturus torosus, undifferentiated (tig-Tg); cross section tlirough gonads, x 188. L 674 -} iiT Plate 3 "J OVUM, CYCLE, AND MENSTRUATION By BERNHARD ZONDEK )sC From the GYNECOLOGICAL-OBSTETRICAL DEPARTMENT OF THE ROTHSCHILD-HADASSAH UNIVERSITY HOSPITAL, AND THE HORMONE RESEARCH LABORATORY OF THE HEBREW UNIVERSITY, JERUSALEM, PALESTINE i OVUM, CYCLE, AND MENSTRUATION ONE FUNDAMENTAL difference between the sexual mechanism in the male and female organisms is of particular importance. While in the male its function is continuous, it is subject to cyclic variation in the female. What are the factors governing these cyclic phenomena? For a long time the theory of the prime importance of the ovum ("Primat cler Eizelle"), advanced par- ticularly by R. Meyer on the grounds of anatomical studies, was generally ac- cepted. This theory implied that the ovary is under the control of the ovum and that the entire generative process receives its stimulus from it; that as soon as fertilization takes place the cycle is discontinued; if, however, the ovum dies without fertilization, the now useless, progestationally developed mucous membrane breaks down and its discharge from the uterus is accom- panied by bleeding (=menstruation). This theory, in our opinion, was rather unsatisfactory for the following reasons: (i) It is based on the fact that the death of the ovum, a destructive process, is the event initiating the onset of the new cycle, the menstrual cycle thus being considered a pathological process which takes place only if the ovum does not fulfill its physiological purpose, fertilization. (2) It is difficult to understand how the ovum, dying on its way through one of the tubes, should be able to set in motion the generative mechanism when one considers that it has no contact with the maternal circulation. After it had been discovered^"* that a gonadotrophic substance is elaborated in the anterior pituitary lobe and that it is possible to reproduce, experimen- tally, the entire generative process with this factor, the role which the ovum plays within this complex was again open to discussion. It is with great pleasure that I am giving an account of these investigations in the Anniversary Volume published in honor of Herbert M. Evans to whom we owe most important progress in endocrinology, including the problems dealt with in the following lines. When attempting to analyze the relationship between the ovum and the hormones involved in the cyclic phenomena, the experimental procedure was to create an opportunity of studying the functions of the ovum independently of hormonal factors and vice versa. 1. What happens if the function of the ovum is eliminated? Ovular function can be destroyed by Roentgen rays. Investigations on this subject (carried out by Parkes^ and at the same time in our laboratory by v. Schubert*^) yielded the following results: Even the application of doses ten times as high as the castration dose (up to 500 r.) does not interfere with the vaginal estrous cycle in the sexually mature mouse, in fact the cycle may continue for weeks or even months. If such animals are sacrificed in mid-estrus, serial sections show very marked changes in the ovaiy, but— and this is the decisive factor— not one single follicle with a large follicular cavity. One finds only remnants of de- [679] 68o Ovum, Cycle, and Menstruation generated ova, but none appear to have survived. The cycle as well as the estrous development of the mucous membrane, therefore, takes place in the absence of functioning ova. Similar results were obtained by giving an irri- tation dose to the infantile ovary (ref. 7, p. 60, 317). Small Roentgen doses (10-25 ^■) ^^^ ^ble to stimulate follicular growth so that the follicle may be developed to the size of a Graafian follicle just before rupturing. No estrogenic hormone is, however, elaborated in these follicles in which growth but no function has been induced. Morphologically, the ova present in these follicles are not to be distinguished from those of normal mature animals. Thus, al- though the follicle is large and contains ripening ova, no estrogenic hormone is produced there and the ripening follicle with its ovum is obviously unable to initiate the production of hormone. 2. Does estrogenic hormone stimulate follicular maturation? This question has to be answered in the negative. Estrus can be produced in infantile ani- mals with estrogenic hormone, but no follicvdar maturation takes place. The gonadotrophic hormone— as is well known— initiates both the production of estrogenic hormone and the ripening process in the follicle. The production of estrogenic hormone is, however, already accomplished (in the twenty-seventh hour) before the growth and maturation of the follicle has even started.*®'^" 3. Does the ovum stimulate corpus luteum formation? Does it produce sub- stances likely to initiate the second generative phase? That this is not the case is clearly shown by the following experiments: (a) In hibernating animals ovarian function slumbers too. It was, however, possible to achieve follicixlar rupture in hibernating bats in winter (ref. 11; also ref. 7, p. 267); in one case pregnancy was obtained in December. (In the bat, as is generally known, copulation takes place previous to hibernation, the living spermatozoa remain- ing in the uterus all the time.) If large doses of prolan are injected into such animals, the formation of several corpora lutea is thus initiated (4-6 in one ovary). (This never occurs under physiological conditions since the bat usually has one young, at the most two.) Large doses may give rise to the formation of corpora lutea atretica in which the ovum is completely pushed aside and, seemingly crushed. No signs of ripening are recognizable in these ova. Not- withstanding the death of these ova, normally functioning corpora lutea ap- pear, as is seen from the effect on the uterine mucous membrane. The ovum cannot, therefore, be held responsible for corpus luteum formation, (b) It could be imagined that the ovum, although dead at this stage, might pre- viously have produced a hormone active in bringing about corpus luteum for- mation. That neither is the case could be demonstrated by Westmann's and later by our experiments'"'^^ conducted on the follicles of rabbits from which the ova had been removed. In our experiments the procedure was to extirpate one entire ovary in sexually mature rabbits, while in the other the ovarian tissue was cut down to such an extent that only a small portion remained, containing just a single follicle on the point of rupturing. This follicle was then punctured and the ovum aspirated. By microscopical examination ol Bernhard Zondek 58 1 the aspirated fluid it was possible to make sure that the ovum had actually been entirely removed. After intravenous injection of prolan this follicle, which contained no ovum, developed into a typical corpus luteum. When, after aspiration of the ovum, the topmost portion of the follicle was also cut off with scissors, even this torso grew into a corpus luteum. These experiments give very clear evidence that a follicle which certainly contains no ovum may be transformed into a corpus luteum in precisely the same manner as is normally the case. From the foregoing evidence the ovum may thus be as- sumed to have no importance for corpus luteum formation. 4. What happens if the gonadotrophic function is inhibited? By giving thallium" to sexually mature mice the cycle is interrupted. In contradistinction to functional interruption, the ovary shows in this case no morphological changes. Large follicles are to be found containing the cumulus oophorus and ova in mitosis. Thallium, no doubt, effects partial inhibition of the gonado- trophic secretion of the anterior pituitary lobe so that, although sufficient amounts of hormone are elaborated to maintain the intactness of the ripening follicle anatomically, the hormone production does not suffice to initiate the elaboration of estrogenic hormone in the theca cells. If now gonadotrophic hormone is injected into the thallium-treated animals they instantly go into estrus. It is, therefore, apparent that, although the cycle has been discon- tinued and the production of gonadotrophic hormone to some extent in- hibited, the ovary nevertheless contains ripe follicles with fully developed ova. That even the complete elimination of hypophysial function has no bearing on the ripening of the ovum is shown by experiments on hypophysectomized animals. H. M. Evans and O. Swezy^^ found, in hypophysectomized but sexually mature rats, the new formation of germinative cells, increased oogenesis, and, in the atresic follicles, the nuclei in the stage of mitotic division with polar bodies. We have examined the ovaries of infantile rats in which the hypophy- sis had been removed on the twenty-seventh day of their life and which had then been sacrificed twenty days later. In these ovaries, too, ripening ova with mitoses were found to be present. There can, therefore, be no doubt that the process of maturation may take place in ova which, for a considerable length of time, have not been under the influence of the gonadotrophic hormone of the anterior pituitary. Summarizing, it may thus be said that the ovum does not stimidate the pro- duction of estrogenic hormone, that estrogenic hormone has no influence whatsoever on follicular maturation or on the growth of the ovtmi itself. The ovum has no influence on corpus luteum formation or the production of its hormone. Neither is there a dependency of the ovum on the anterior pituitary and its gonadotrophic hormones. It thus appears that the ovum leads its own life, its development does not depend in any way on either the follicle Avith its hormone or the anterior pituitary lobe. The ovary is the host of the ovum but at the same time its servant, since with its hormones it assists in preparing the uterine mucous membrane anatomically and functionally for the rcceji- 682 Ovum, Cycle, and Menstruation tion of the ovum after fertilization. Nevertheless, the anterior pituitary is the superior factor in this process. Without the anterior lobe, without gonado- trophic hormone, no follicular rupture would be possible, and the ovum could not migrate through the tube to fulfill its natural function, fertilization. The ovum itself, therefore, is of no significance for the process of hormonal pro- duction in the ovary nor for the growth and breakdown of the uterine mucosa. It has no bearing on cyclic changes. Segmentation of Ovum Although this problem is not directly related to the menstrual cycle, we are taking the opportunity of referring briefly to our experiments concerning the segmentation of the ovum. The subject usually employed for the study of this phenomenon is the pregnant animal, since in its ovary follicular atresia is particularly marked, and segmented ova are preponderantly met with in atresic follicles. Spuler^'' holds the view that, in the degenerated granulosa cells, substances may be formed which induce segmentation in the ovum. L. Loeb"'^^ holds changed chemical and inechanical conditions in the atresic follicle responsible for this event, while Haggstrom" takes changed osmotic pressure or difficult supply of nutritional factors to the follicle into considera- tion. Is segmentation actually nothing but the consequence of follicular atre- sia? In order to study this question (ref. 7, p. 329) it was necessary to produce artificial follicular maturation in the rat by means of prolan injections, in order to find out whether in these ripening follicles segmentation might also be discovered. A total of eighteen infantile rats were treated with prolan, and serial sections of their ovaries were examined microscopically. In thirteen of the animals segmented ova were in fact encountered. It seems to me note- worthy that these segmented ova were present not only in atresic follicles but also in the ripening ones as well as in those that had already reached the stage of vascularization and, in part, luteinization. In one rat which had been treated with prolan a segmented ovum was found on its way through one of the tubes after follicular rupture (pi. 1, fig. 1). This tubal ovum consisted of five segments of different size containing altogether ten chromatic bodies. Some granulosa cells were still adhering to the ovum. It may thus be assumed that segmentation is not a consequence of atresia. Is the segmentation of the ovum to be considered a regressive process? We do not think that this is the case. Two separate processes have to be taken into account: (a) Since the ovum shows signs of maturation, and, moreover, seg- ments of equal size with perfect nuclei and mitoses are present, a mitotic divi- sion of the ovum must be assumed; (b) even more frequently than this division another picture presents itself which, in our opinion, is due to the chromo- somes being drawn apart and secondarily enveloped in plasma so that seg- ments with irregularly distributed chromatic bodies are formed. That segmentation of the ovum does not depend on the anterior hypophysis is shown by the fact that a similar occurrence is also met with in infantile Bernhard Zondek 680 hypophysectomized rats. Plate i, figure 2, shows an ovum tousisiing of five seg- ments in a follicle of a rat (forty-seven days old, hypophysectomized twenty days earlier). Cell no. 1 contains a nucleus, cell no. 2 a nuclear spindle; in cell no. 3 only a suggestion of the small nucleus is recognizable in the photograph; cell no. 4 contains no nucleus; cell no. 5 shows cytolysis. Plate 1, figure 3, shows three segments of the ovum in a follicle of medium size of an infantile hy- pophysectomized rat. Two of the segments contain nuclear spindles, the third a polyaster. This may possibly indicate that subdivision has taken place. In the segments, vacuoles are clearly seen, as a sign of the early stages of degeneration. Ovular segmentation is, therefore, not a consequence of follicular atresia since it occurs in the ripening as well as in the rupturing follicle. We are inclined to assume that this phenomenon must be considered a degeneration on parthenogenetic lines. The ovum which very early (even in the infantile ovary) displays its ripening tendency may continue its development even with- out fertilization. This parthogenetic impulse ends in an abortive effect, since very soon degeneration sets in and the ovum dies with signs of cytolysis. The constructive process is, therefore, very soon replaced by a degenerative one. Theory of Cyclic Menstruation In the foregoing we have seen that the ovum leads its own life in the ovary and is of no significance to the cyclic process and the initiation of menstrua- tion. In the following lines a brief account will be given of how the cycle and the appearance of menstruation can be imagined to be brought about:'" Men- struation is a complex process initiated by the functional, harmonious coop- eration of the anterior lobe of the pituitary gland on the one hand, and ovarian hormones and uterine mucosa on the other. The regulating factor of the en- tire cycle is the estrogenic hormone— the estrogenic hormone level operating similarly to the mercury level in the regulator of the thermostat. The cycle can be imagined to run as follows: In the theca cells, constantly increasing amounts of estrogenic hormone are elaborated, the peak produc- tion being in the middle of the cycle. The high estrone level cuts off prolan A and this initiates the activity of prolan B. This hormone brings about follicular rupture, corpus luteum formation, and progesterone production. Since now. however, prolan A is no longer active, the estrone level is lowered, and this lowering starts again the activity of prolan A, so that estrogenic hormone is again produced. At the peak of production, that is, when the corpus luteum is fully developed, the very high estrone level cuts off both gonadotrophic factors (A and B) in the anterior pituitary. The gonadotrophic hormone can no longer enter the blood stream, and in this way both progesterone and es- terone levels are lowered. Some days later the bleeding sets in. With the lower- ing of the estrone level, prolan A automatically comes again into activity, previous to and during the hemorrhage, and the cycle of follicular maturation is again begun. In brief: (1) lowering of the estrogenic hormone level brings prolan A into activity, (2) high estrogenic hormone level cuts off prolan A, 684 Ovum, Cycle, and Menstruation and (3) very high estrogenic hormone level cuts off both prolan A and B and later initiates menstruation. We do not know what happens in the interval between the exclusion of prolan A and B and the resulting lowering of the estrone and progesterone levels, on the one hand, and the bleeding, on the other,— that is, the actual in- itiator of the bleeding is as yet unknown. The experimental induction of uterine bleeding through esterone and/or progesterone suggests the existence of a chain reaction initiating the activity of a special bleeding factor which may be produced in the organism by the hormones themselves. Further ex- periments will have to prove whether this hypothesis can be confirmed. REFERENCES 1. Zondek, B.: Deutsch. med. Wchnschr. 52:343, 1926. 2. Zondek, B.: Ztschr. f. Geburtsh. u. Gynakol. 90:378, 1926. 3. Zondek, B.. and Aschheim, S.: Arch. f. Gyniikol. 130:1, 1927. 4. Smith, P. E.: Proc. Soc. Exper. Biol. &: Med. 24:131, 1926. 5. Parkes, A. S.: Proc. Roy. Soc. London, s.B 101:71, 421, 1927; 102:51, 1927. 6. V. Schubert, E.: (Habilitationsschrift, Univ. of Berlin, July 1926). 7. Zondek, B.: Hormone des Ovarhuns iind des Hypophysetivorderlappens (2d ed.; Wien: 1935)- 8. Zondek, B.: Jl. Endocrinol. 2:12, 1940. 9. Zondek, B., and Sklow, J.: Jl. Endocrinol, (in press). 10. Zondek, B.; Sulman, F., and Sklow, J.: Jl. Endocrinol, (in press). 11. Zondek, B.: Lancet 231: 1256, 1933. 12. Westman, A.: Arch. f. Gynakol. 156:550, 1934. 13. Zondek, B.: Jl. Physiol. 81:4, 1934. 14. Buschke, A.; Zondek, B., and Bermann, L.: Klin. Wchnschr. 6:683, iQ^?- 15. Evans, H. M., and Swezy, O.: Mems. Univ. California 9: 1 19, 1931. 16. Spuler, A.: Anat. Hfte., L Abtl. 16:85, 1901. 17. Loeb, L.: Arch. f. mikroskop. Anat. 65:728, 1905. 18. Loeb, L.: Ztschr. f. Krebsforsch. 2:259, 1912. 19. Haggstrom, P.: Act. obstet. et gynecol. Scandinav. 1:137, 'Q^S- 20. Zondek, B.: Clinical and Experimental Investigatioris on the Genital Functions and their Hormonal Regulation (Baltimore: 1941), 243. EXPLANATION OF PLATE PLATE 1 Fig. 1. Microphotograph x 213. Tubal ovum of an infantile rat treated with prolan. Five segments of the ovum with granulosa cells adhering to them. Fig. 2. Microphotograph X 473. Five segments of an ovum in the follicle of an infantile hypophysectomized rat. Cell no. 1 contains a nucleus, cell no. 2 has a nuclear spindle, cell no. 3 has the shadow of a nucleus, cell no. 4 is without nucleus, cell no. 5 is cytolytic. Fig. 3. Drawing x 790. Three segments of an ovum in the follicle of an in- fantile hypophysectomized rat. Two of the segments contain nuclear spindles, the third a polyaster. [686: m«'M^ "^ m 1 ,v-< :*f;* .•*. arj^s Kt m- T-? ^ N ^ -»i^'^ Plate 1 i 1 I :-":^l