S I X FIGIJRES Tlic pituitary gland of tclcosts has received scant attention. Sifc~nclcll ( ’14) des(~ri1)osthe relatioiiships and striictnrc of this organ in scvcral of the bony fishes with special rcfcrciicc to ilnguillu, E:POS and (:yprinns. tic. Beer ( ’26) gives R hrief dc script ion of thc I)i tiii t ary of t clcosts in general, emphasi zing the fact tliat little is Iinown of the structure of the gland in tliis group, and more remn E’loreiitiii (’34) has giveii a short account of the pituitary of teleosta. While carrying out some experiments 071 the sex cyale i n Fiindulus it was foi11-d necessary to analyze the structure of tho pituitar;v of this animal. (:lands of 1ij2 animah have so far hc?n PXamiried, of which sc\-eiity-cigIitwere untreated norrrial animals, iii(1lutling Imth iiialcs and femwlcs, and form the basis for the followiiig tlcw4l)tion. Thcsc animals were collected at various times from the l)elaware River. Some were killed irniiieditttely, others werc kept f o r longer or shorter periods iii the laboratory and iiidividualx killctl ;It definite intervals. A few of the animals, t h o s ~killed in July, were froni ’iVoods IIolc. For microscopic study aerial sections were madc and stained with amr-cosiri, the stain found to be the most satisfactory of several that were tried, sections rcinaiaiiig iu the stain for 01115.- 4 hours. ‘l‘lic pituitary gland of E’uiidulus lies on thc ventral surface of the brain immetliaiely postel-ior t o the optic chiasma and betweeii the iiiferior 1ol)es. It is covered velitr~llyby the n:ir~~sphcnoiclhone although there is no excavation in this 357 1)orie cornparahlo l o :L s e l h tiiwica. ‘I’li~ gland caii be swn l.(wtlily Iliroagh the i*oof of the mouth a s iooii a s ilie iriiicoiis membrane is dissectcd an.ay from thci micl-line a i d mav be reriiovetl froni survival aniirials by this appi*owcli ( 11atthcn -, ’ 3 3 ) . The infiinclihalar stalk is short arid relatively broad, ruiining ventrally from the clienceplialon l o the inccliaii portion of the pituitary, IVhcii exarriiiied from the vcliitral view in a fresh cvxiditioii I h t t gland 1)imeiits three divisions, a larger ior and a sina1Icr posterior pari of compact millt;r.-~vhi te tissue, t l i ~ s ctwo parts being separated by u riarrow I ~ ~ C ~ U I R T zoiic of clearor material. Tlie significance of thwe surlacc marlrings api)ears wlien scxtioiis of the gland are c~xuiniiietl. 1wo iriaiii rogioiis m a y then he ilistinguislied, a pal’s ~ ~ C L ’ V O F B aiid a pars glanclnlaris, the latter being further siibtliuitlctl into a so-callcd iransitional rcgioi? and a pars intermedia (fiz. 1). Tlir?rc is no hj-popliysial c:uvit y. The pars i i e r v o s ~ , as i n vcrtcbralcls in p r i e m l , is composed mainly of fibrous ~iiatcrial. Fibers lrom t l i ~infundihulum extend into the gland from the median rloiml point where ihc: iiifiintlibnlar stalk joiiis the pituitary arid from this region radiate out in a11 directions into t hc siihslaiice of the pars glandularis. In the a i i t P r i o r portion of tlir gl!:lwntl11ies~fibers are sc1)arated f r o m ihe free snrface by a r ~ l a l i r c l ywide zo1ie of glaiiclidar cells, but a t one region, sonicwliat iicarcr the posterior end of tlic pituitary, the fihcrs of ihc pars iicrvosa reach quite mar r 1 t’ig. 1 Fundillus pituitar?, sagittal section. x 66. PS, pars ncrvosii; 1’1, pars intermedia, and PT, traiisitiimtl Icgion. the two sul)tiivision.i of t h r par‘ glnndiilnris. Fig. 2 Acidnphilicz erlls from traiisitional rtgioii of ariirrial killrd 12/17/32. granultv stain less intcnsc4y tEiaii (‘amera I i ~ i d outlinr, s x 900. In these c d l s do those iii the acidophilrs of the pars interm a (fig. 5) and do mot photograp11 \ve11. Fig. 3 Cllroniophobic cells from transitional rcBgion of ariim:tl killrcl 7/30/33. Canicm lucida outline, X 900. Fig. 4 D e ( ~ p t a i n i a gbayophilcs f r o m posterior pole of animal killcd R ’24,’ X 690. Fig. 5 Par? intrrmrilia killccl 9/54/35. X fi90. A(‘, acidnphile; (’11. chroiiro phobc ; PS,prowss of pars m r i 088. Fig. (i Pars interinedia of : m i n d killcil 10/9/35. X 690. l:, basophilc: (‘TI. vhroniophohe ; PX, process of pars ncnos:i. Figures 1 t o 6 359 the ventral surface. Tt is this zone wliich can be seen in the fresh coiitlitiori as a narrow irregular band of clcarer t issue sq)aratiiig the anterior two-thirds of the gland from tho posterior third, N S mentioncd above. Reattcred among thc fibers of the pars n(?rvosa are a number of neurogliu cells. Large masses of colloid material may be seen only occasionally, but considerable amounts of small droplets or graiiules can be foinnd scattered freely among the fibcrs. This mattlrial is acidophilie and is most abundant n r w thc pars intermedia which borders the pars nervosa. Stendell believes tliat these arc droplets of secretion from the cells of the pars intermedia. Thc similarity in staining between these granules or clroplcts and the cptoplasmic inclusions of ilic acidophilic cells of the I ) N ~ S intermedia immediately adjacent to the pars n e r ~ o ~ a suggwt s a relationship between the two in Fundulixs idso. The pars g l i ~ ~ t l u l a r of i s tclcosts lias been subclivicl(x1 by Stendell aiitl by dc Rccr inlo thrcv rcgions: il pars aiiterior forming a small aiitero-dorsal portion ; a pars intcrmectia c o n sistirig of hasophilic 0011s huiiding the processes of thc pars nervosa ; arid a iraiisitionul regioti (uebergtingsteil) which forms the hulk of this part of the gland. In Fimdulus, however, thc cells in the antero-dorsal region are so likc those of thc rest of this part of the gland that 110 pars anterior comparable t o tliat descrihetl f o r other teleosts can 1)c disuished. The entire region forruing the anterior half of the pars glandularis and cxtencling posteriorly and laterally corresponds in position arid struetiire to what Stendell calls the transitional portion (fig. 1). It is composed chicfly of acidophilic cells with occasioiial basophi les ailti chromophohes. The acidophilcs vary in appcitranc~i i i animals killed at tliffercnt periods. T n general they are large cclls with nuclei mhich arc usiially carescent-shaped or folded, rarely spherical, the cytoplasm containing h e , uniformly tlistrihinted granules which arc weakly aciclophiliv. In ariiinals killed in late summer and fall, howcw?r, tlie granules are iargcr, l t m regular iii their occiirrenccl and stain more intensely (fig. 2). Tlie relatively rare haso1)liiles rewmble the acidophiles iii shape and niiclear Y t r uc t u re . PITUITARY GLAND OF FUNDULUS 361 I n addition to the so-called transitional region, the pars glandularis-presents a pars intermedia, which is made up of a zone several cells in thickness bordering the processes of the pars nervosa (fig. 1). I n the anterior part of the gland the pars intermedia is separated from the free surface by the transitional portion. At the posterior end, however, the transitional region is interrupted so that the posterior pole of the gland is formed entirely by the pars intermedia. The cytoplasm of the cells of the pars intermedia is usually basophilic. Chromophobes are found occasionally, however (fig. 6 ) , and in many animals those cells which are immediately adjacent to the processes of the pars nervosa are strongly acidophilic, the granules in the cytoplasm staining a deeper red than do those of the acidophiles in the transitional portion (fig. 5). These acidophiles are always somewhat larger than the basophiles or chromophobes. In addition there may usually be found a number of large, sharply outlined cells of which the cytoplasm stains a pale lavender with azur-eosin. These cells are particularly noticeable at the posterior pole. I n general they appear like somewhat faded acidophiles. Finally, in addition to the ordinary type of basophile found in this region, in which the granules are small and regularly distributed throughout the cell, a striking feature of the pars intermedia of animals killed at certain periods is the appearance of large cells with irregularly distributed masses of very deep staining, basophilic material. The nucleus of this last cell type is large, pale with a prominent nucleolus, and is often eccentric in position (fig. 4). The pars nervosa and the transitional region of the pars glandularis are well supplied with blood vessels. The pars intermedia contains no blood vessels, though its cells are all near the abundant vascular supply in adjacent processes of the pars nervosa. I n the series of animals so f a r examined the cells of the pars glandularis appear to undergo cyclic changes. In the transitional portion of animals killed at different periods differences were observed in the relative numbers of chromophobes and acidophiles. I n seventy-one animals killed from T H E ANAMMICAL RECORD, VOI.. 6.5, NO. 3 362 SAMUEL A. MATTHEWS September to Map the cells of this region were almost entirely acidophilic with only occasional basophiles or chrornophobes In the early part of this period, from September to December, the acidophiles were large with prominent, deep-staining granules (fig. Z ) , while in animals killed from January to March, the acidophiles were somewhat shrunken and stained less intensely. The most striking difference in the appearance of this region, however, was shown during or at the end of the breeding season, which extends from Xay until the latter part of July. I n seven animals killed in late May and July the transitional region of the pituitary consisted almost entirely of shrunken, chromophobic cells (fig. 3). From an examination of these animals it would thus appear that the transitional region of the gland is predominantly chromophobic in the summer months, acidophiles appearing in the fall and persisting throughout the winter, though staining less intensely in late winter and spring. In the pars intermedia differences were also observed in different animals both in the number of acidophilic cells adjacent to the pars nervosa and in ihe presence or absence of those basophiles with large, deep stzining granules described above. Acidophiles varied considerably both in numbers and in staining re8ction.l I n the cells counted the percentage of acidophiles was found to be consistently high during May, July, and September, varying fro= 50.3 to 55.8 % with an average of 53.4. While as a rule these cells were less numerous in the fall and winter months such wide variation existed Cell counts of this region mere made by the sampling method. A ruled square mounted in the ocular was focusscd in such a way that as much of one side of the square as possible rested on the boundary between pars nervosa and pars intermedia and in a11 cases the enclosed area mas selected 80 that the maximum iiumber of acidophiles was included. A count way then made of acidophilea against the total number of cells in the circumscribed area, using every fourth section. The figures thus represent the percentage of acidophiles in the number of cells counted and not the percentage present in the pars intermedia as a whole. The latter figure would be considerably lower in all cases owing to the fact that acidophiles are found only along the border of the pars nervosa, the region chosen for making the count, while below this region are several layers of basophiles or chromophobes. PITUITARY GLAND OF FUKDULUS 363 that any generalization concerning this period is difficult to make. Thus in October the acidophiles averaged 34.5 "/. of the cells counted ; in December 40.7 ; in January ( '35) 34.9 "/. (weak-staining acidophiles) ; in February 32.4 % ; while in March and April four animals showed no acidophiles whatever in this region. On the other hand, in January 1936 the acidophiles averaged 49.2 %, o r nearly as high as during the previous July, and the cells were heavily stained. Before they were killed these animals must have been subjected to a variety of conditions with respect t o food supply, temperature of the water, etc. Individual differences such as those observed in the animals killed in January 1935 as compared with those killed in January 1936 may be dependent on differences in the previous history of the two groups of animals. At any rate while the number of acidophiles in the pars intermedia varies, in general being highest in late May, July and September and lowest in March and April, the variation in individuals killed in the same month of successive years indicates that factors other than seasonal changes alone must bc concerned in this variation. In addition to these changes in the acidophiles of the pars intermedia the occurrence in this region of cells containing large granules of deeply staining Fasophilic material varies considerably. These cells are particularly numerous, when they occur at all, at the posterior pole of the gland. Apparently they may be found in individual cases throughout the year but they are much more common during summer and fall than in winter. Thus in all animals killed during July and September these cells were found in large numbers and in October they occurred in moderate numbers in four out of the five animals killed during this month. I n December, however, only seven out of twenty-three animals showed these cells and from January through May they were even rarer, occurring in only five out of thirty-eight animals killed during this period. What relation these various cells of the pars intermediachromophobes, basophiles, acidophiles and cells with large THE ANATOMICAL REWED, VOL. 65, NO. 3 364 SAMUEL A. MATTIIEWS basophilic granules-bear to one another is difficult even to estimate from a study of these preparations, and of course conclusive evidence of any relationship, such as might be obtained if the cells could be observed in the living animal, is impossible in material fixed at different periods from different individuals. No cells intermediate in type between chromophobes and chromophiles or betwccn basophiles and acidophiles have so far been found. The only evidence suggestive of relationship lies in the fact that deep-staining acidophiles present in some animals appear as a single layer of cells separating the rest of the pars intermedia from the pars nervosa, yet in the same region of other animals only basophiles or chromophobes can be found (figs. 5, 6). Obviously the acidophiles must develop from either the chromophobes or the basophiles but so far no evidence has been obtained as to which is the case. I n the teleosts which he examined Stendell believed the basophilic granules were ‘unripe, ’ acidophilic granules representing ripened secretion. On the other hand the recent work of Severinghaus (’33) indicates that in the anterior lobe of the mammalian pituitary gland both acidophilic and basophilic cells originate independently from chromophobes, the chromophobic antecedents of the two types being distinguished by differences in their Golgi apparatus. Such work indicates the caution that must be observed in drawing conclusions from the staining reaction of granules in the cell alone. From this examination of these seventy-eight animals it seems fair to conclude that the pituitary of Fundulus is not static but that it undergoes cyclic changes. With what these changes are associated, however, it is difficult to say, particularly since the functions of the pituitary in teleosts is not well known. A few reports indicate that it may be concerned in some way with the sex cycle. Thus Houssay (’30) injected seven South American teleosts (Cnesterodon) with saline suspension of the pituitaries of four large shad on September 27th. Three of these expelled eggs on the twenty-eighth and a fourth on the twenty-ninth, the other three animals PITUITARY GLAND OF FUFDULUS 365 dying. Similar results were obtained with another animal which expelled eggs on the thirtieth of September, 2 days after injection. Control injections of muscle extracts were ineffective. This season is apparently near the normal breeding season for this species, however, for he found it necessary to discontinue the experiments 03 the eighth of October, since at this time several animals expelled eggs spontaneously, a fact which makes it somewhat difficult to estimate just how important the injections were in obtaining expulsion of eggs. More recently Cardoso ( '34) injected saline suspensions of the pituitaries of Pimelodus into other individuals of the same species and obtained striking increases in weight in both testes and ovaries. These experiments suggest that the pituitary gland plays some part in control of the sex cycle in this group of vertebrates. In addition it has been shown that in some teleosts at least, notably Phoxinus and Ameirus, the pituitary is concerned in part with the pigmentary effector system (Giersberg '32; Parker, '34). While in Fundulus it is not essential for ordinary color changes (Matthews, '33) the pituitary of this animal does contain a substance capable of producing dispersion of t.hc melanophore pigment in the catfish, frog and lizard (Rleinholz, '35). The melanophore principle is produced by the pars intermedia of other animals (frog, Allen, '30; rat and mouse, Geiling and Lewis, '35). Hence it may be significant that the highest percentage of deep-staining acidophiles occurs in the pars intermedia of Fundulus from May to September inclusive, since Parker and Brower ('35) have shown that a nuptial sex character consisting of a well-developed black spot on the posterior part of the dorsal fin is present in Fundulus at approximately the same time, namely from April to November, but not at other times of the year. While similar nuptial characters in other teleosts are associated with the gonad ( Courrier, '21 ; Blacher, '26; Kopec, '27; Bock, '28; Tozawa, '29) the gonad may in turn be controlled by the pituitary. As far as the differences observed in the transitional portion of the pituitary at the end of the breeding season are concerned, the lack of acidophiles and large numbers of chromophobes at this period 366 SAMUEL A. MATTHEWS may be associated either with sexual exhaustion or with general metabolic changes that may occur a t this time. During and after the breeding season Fundulus is more susceptible to handling and disease that at any other period. I n this connection there may be mentioned the following observations. I n a series of animals which were injected with sheep pituitary extract the concentration employed was obviously toxic due probably to the phenol with which it was preserved. The animals were sluggish, refused food, and when they were killed their peritoneal cavities were found to contain considerable quantities of exudate. Seven of these cases have been observed and in all of them the transitional region of the pituitary showed few if any acidophiles, consisting almost entirely of chromophobic cells. I n fifteen controls, however, the same region of the pituitary was strongly acidophilic. Briefly, then, in a series of seventy-eight normal Fundulus it has been found that the glandular portion of the pituitary shows certain cytological differences in animals killed at different periods, differences which are thought to represent a cycle of changes that occurs in each individual. With what these changes are associated can be determined accurately only by appropriate experiments. LITERATURE CITED ALLEN, B. M. 1930 Source of the pigmentsry hormone of the amphibian hypophysia. Proe. SOL Exp. Biol. and Xed., vol. 27, p 504. DE BEER,G. R. 1926 The Comparative Anatomy, Histology, and Development of the Pituitary Body Oliver and Boyd, London. BLACHER, L. J. 1926 The dependence of secondary sex characters upon tcstieular hormones in Lebistea reticulatus. Biol. Bull., vol. 50, p. 374. BOCK,I?. 1928 Kastration und sekundire Geschlechtsmerkmale bei Teleostiorn. Zeit. f. wissensch. Zod., Ed. 130, 8. 455. CARDOSO, D. 1934 Relations entre l'hypophyse et les organes sexuels e h a lee poissons. Compt. Rend. de la SOC.de Biol., T. 115, p. 1347. C ~ V R R I F ~R., 1921 8ur le conditionncment des caracthres scxuels secondairos ehez les poissons. Compt. Rend. de la Soc. de Biol., T. 85, p. 486. FLORENTIN, P. 1934 Histophyaiologie compar6e do 1'hypophysoe. L 'excr6tion de la oolloide hypophysaire chez les t616ost8ons. Ann. de Physiol., T. 10, p. 963. PITUITARY GLAKD O F E’UXDULUS 367 GEILING, E. M. AND M. I(. LEWIS 1935 Further information regarding the melanophore hormone of the hypophysis cerebri. Am. J. Phys., vol. 113, p. 534.. GIERSBERG,H. 1932 Der Einfluss der Hypoplryse auf die farbigen Chromatophoren der Elritze. Zeit. f. vergl. Phys., Hd. 18, 8. 309. IIOUSSAY, B. A. 1930 Accion sexual de la hipofisis en 10s peces y reptiles. Revista de la Sociedad Argentina de Biol,, vol. 6, p. 686. KLEINHOLZ, L. H. 1935 The melanophore-dispersing principle in the hypophysis of Fundulus heteroclitus. Biol. Bull., vol. 69, p. 379. KOPEC,S. 1927 Experiments on the dependence o f the nuptial hue on the gonads in fish. Biol. Oen., Bd. 3, 9. 259. MATTHEWS,S. A. 1933 Color changes in F’unciulus after hypophysectomy. Biol. Bull., vol. 64, p. 315. PARKER, G. IF. 1934 Color changes of the catfish Ameiurus in relation to neurohumors. J. Exp. 2001, vol. 69, p. 199. PARKER, C-. H. AXD H. P. B R Q ~ E R 1935 A nuptial secondary sexcharacter h Funundulus heteroclitus. Biol. Bull., vol. 68, p. 4. SEVERINGHAUS, A. E. 1933 A cytological study of the anterior pituitary of the rat, with special reference to the Golgi apparatus and t o cell relationship. Anat. Rec., vol. 57, p. 149. STENDELL, W. 1914 Die Hypophysis Cerebri. I n Oppel’s “Lehrbuch der vergl. mikros. Anat.,” Teil 8. TOZAWA, T. 1929 Experiments on the development of the nuptial coloration and pearl organs of the Japanese bitterling. Folia Anat. Japon., Bd. 7, 9. 407.