DISSECTIONS O F HUMAN SEMINIFEROUS TUBULES FRANKLIN P. JOHNSON Departments of A n a t o m y and Urology, University of Oregon Medical SchooT ELEVEN FIGURES Whether the tubules of the testis terminate blindly or in the form of anastomosing loops has for many years been a matter which has never been satisfactorily determined. This is attested t o by the fact that in our numerous textbooks of histology the authors a r e a t variance in regard to the form of the seminiferous tubules, and, furthermore, by the fact that none of the authors are able to represent the tubules graphically in any but the most diagrammatic manner. When one considers that such eminent anatomists a s Sappey, Hyrtl, Eberth, Kolliker, Krause, Bremer, Huber and Curtis, and others have studied this problem and have drawn conclusions which a r e a t variance with one another, he may correctly conclude that there are difficulties in its solution which are not easily overcome. The chief difficulty lies in the extreme length and tortuosity of the seminiferous tubules, and in their closely packed condition. The attempts to solve this problem have been chiefly by means of dissections of acid macerated material made under the dissecting microscope. However, Bremer’s study of the human fetal testis was made from serial sections with the aid of wax models, and Curtis made a study of the tubule of the rat’s testis by the same method. The difficulties of the modeling method, its tediousness, and the high possibility of error, have discouraged and prevented its successful application to the adult human testis. Maceration and dissection, therefore, still remain the method of choice for studying the morphology of the seminiferous tubules, its success, in large part, depends upon proper technique. 187 188 FRANIi1,IN P. J O H S S O N The method of maceration used in the present study is quite siniilar to that which the authoiu used in 1916 in the separation of tlie lolmlcs of the pig’s liver. The testes are first thoroughly hardeiiecl in foimalin in toto. Segments ahout 1 cm. iii thickness a r e tlieii iiiatlle, the cuts l x i i q l ) a i ~ ~ l ltoe ltlic. direction of the scptala or lobules of the testis. The ring of tunicit vaginalis si~r~~ouiicliiig the seginent slioulcl not be removed but it should he cut through in oiic o r two places with ii i-azoi*. Each segment or lnlock is then submersed in i,5 p e r c w i t hydrochloric acid i i i ~ v h i ~itl i is allomed to remain for 1 (lay to 1 week. Tt is then placed in a suitable dish for clisseetioil, preferably a low Stendcr dish, compleiely covered by the acid. Heat is then al>l)liedhy means of a Runsen burner and the acid slowly raised to a temperature just short of hailing. A s the temperature rises one notes a niai*ked slirinkiiig of the tunica albuginca, which, unless the ahove-described cuts have been niade, greatly w i r p s the segment. With contiiiuecl heating the parcnchyina it self shrinks, and a n y warping which may have occurred corrects itself. The amount of shrinkage is very great, at least 30 to 40 per cent by volume, but as the shrinkage is uniform, neither the continuity of the tubules nor their form is seriously disturbecl. A s the macerat ion progresses the s ~ e c i n i c nt~ iriisa dark brown color nil hecomes quite soft. The heat is now tipplied very slowly so a s to eliminate ‘bumping,’ as a single ‘bump’ will completely ruin the sljecimen if it occurs as the maceration nears completion. The length of time required by the heating pi*ocess is froin 20 to 30 minutes, depending upon various factors. The optimum degree of maceration is reached when the collagen fibrils a r e completely dissolved, and appear about the specimen as a dark brown sediment. Thc dish is then removed from the Ruriseri burner. care being taken to avoid rough handling. A portion of the acid is removed with a pipette and water is added, and this pi-occss repeated until practically all of the acid has been iwnoved. T t is important that boiled water be used f o r this purpose as hoiling clispcls the a i r and prevents the formation of minute a i r bubbles which cling DISSECTIONS OF H U M A N SEILIINIFEROUS TIIBUI,ES 189 to the tubules and greatly interfere with successful teasing. The water turiis the tubules a yellowish-white color and the specimen is now ready for the dissecting microscope. If the tubules cannot be lifted away from one another with great ease, the maceration has iiot been sufficientlr complete; on the other hand if the tubules b m i k when an attempt is made to tease them, the niaccratioii has been carried too far. In Fig. 1 IIuman seminiferous tubule of the blindly ending type. The small straight tubule is a t the lower end of the figure. Tlie blind end is located about half-way between the rete and the periphery of the testis. F o u r blindly ending diverticula are s w n a t I). A n uncommon type. Fig. 2 Siniplc loop type of seminiferous tubule, there being no anastomotic branches present. This is also a n uncommon type of tubule. Fig. 3 The two anastomotic branches at the lower end of the figure form a complete ring of convoluted seiriiniferous tubule (see also fig. 9). Such rings are likewise rare; another is sliown on tubule 13, figure 11. Fig. 4 A single loop bc+vccn two near-adjacent seminiferous tubules. I n this and subsequent figures; I), diverticulum; B, branching, X tubule broken or not followed f o r other reasons. 190 F R A N K L I N P. J O H N S O N either event it is wise t o discard the specimen and begin over again, as the best dissections cannot be obtained from such material. Success with this method depends upon the proper degree of maceration, and the determination of this requires experience. In the present study observations were recorded by means of a series of drawings which were made as the dissections Fig. 3 Three seminifcrous tubules with anastomosing branches forming two complete loops and a complete ring. Fig. 6 Four tubules forming three eompletcly dissected loops. progressed. Each branching of a tubule was scrutinized most carefully by rolling it over and over to exclude the possibility of a broken tubule being adherent to the tubule which m7as being followed. Likewise each tubule ending was examined carefully from different angles to determine whether it was a rounded blind end o r the end of a cut or broken tubule. It DISSECTIONS O F H U M A N SEMINIPE1:OVS ‘IY!BUIXS 191 was found impossible to record all the convolutions of the tubules as they exist in their three dimensions, a s certain portions of the tubules overlie one another and the branchings and other features of the tubules cannot be properly shown. They were, therefore, drawn a s though they existed in one plane and the frequent crossings, overlappings and change of Fig. 7 The small loop towird the center of the figure does not reach half-way t o the periphery of the testis. The nuiiierous divertienla give the impression of blind ends of tubules with anastomoses between them. plane were largely eliminated to make plain what the dissection revealed. The direction and approximate length of the tubules and the number of branches and diverticula were represented a s accurately a s possible. Photography was found of little value in the recording of observations. I n order to photograph a tubule it is neces- 192 FRANKLIN P. J O H N S O R sary to remove it from the underlying tubules to a black background and fix it in mounting gelatin. Sniall portions of a tubule can be so handled, but the larger loops and series of loops present difficulties so great that it did not seem worth the eEort. Moreover it was found that because of their elasticity, the tubules refused to lie in a position which would Fig. 8 Another group of four tubules forming three coiiipletc loops. permit the clear demonstration of branching, diverticula, anastomoses, etc. When the elastic tissue surrouiiding the tubules was softened or dissolved by digestion in artificial pancreatic juice, the tubules became so fragile that they could no longer be handled. Dissections should be begun at the straight tubules. The uppermost lobules should be carefully lifted up and discarded as these will contain iiumerous cut ends of the tubules. Such DISSECTIONS OF HUMAN SEMINIFEROUS TUBULFS 193 dissectioii of whole lobules is easy towaid the mediastiiiuni where the lobules a r e me11 marked, but becomes increasingly more difficult toward the periphery of the testis, as in this region the lobules are frequently ill-defined and a r e held together by those tubules wkicli pass from oiie lobule to another. Fig. 9 l’liotograpli (retonclicd) of :I complete ring of convoluted tubule. This ring is also shown in figure 3 ;uld is oriented the same \ray. The t w o straight tubules and the sliort low(^ anastomotic tuhulr s l l o ~ n at the bottom of the figure :ire coiled togctlicr in a nmss t h t rcquired dissection to dc~iioiistrntc tlicir form. The actual teasing of p r o l m l y l)repared speciiiiens is not a difficult procedure. The acid dissolves the collageii fibrils and leaves oiily a frail reticulum wl~icllbinds the tubules together. The elastic fibrils, however, a r e riot destroyed and retain the greater part of their elasticity. These fibrils, as sliowii by Hill form a close network around each tubule but they do not hind the tubules together. T i 1 teasing, therefore, it is only necessary t o lift the tubule from its bed by breakiiig the 194 F I i A N R L I N P. J O H N S O N reticulum, while the elastic network about it gives it great tensile strength. I t is necessary in following a tubule to pull it out sufficiently straight with the dissecting needles to definitely establish its continuity, and to scan it for diverticula and branches. Upon being released, however, it largely assumes its fownier shape. As described by numerous investigators, tlie straight portion of the sciniiiiferous tubule is about one-half the diameter of the contorted portion, is thin-walled and quite straight. It attaches at one erid to that plexus of epithelial sinuses known as the rete testis and at the other to the contorted tubule, where the deniarcation is verS abrupt. A t times one sees a caecumlike ending of the contorted tubule, the straight tubule joining its side. Such caecum-like terminations may be eloiigaled to form direrticula as shown in figure 11, tubule 12. F o r tlie most part the straight tubules al*e short a i d of approximately equal length, so that oiie finds the majority of the terminations of the contorted tubules equidistant from the rcte testis. Occasionally, however, one finds a contorted tubule which falls short of reaching this position, and its straight tubule, two or three times the ordinary length, will be found coursing alongside the DISSECTIONS O F HUMAN SEMINIFEBOUS TUBULES 195 terminal portions of other contorted tubules. Straight tubules usually reach the rete singly, although occasionally two may unite to form a comiiion trunk (fig. 11, tubule 8). If a single contorted tubule is followed from the straight tubule peripherally, it is noted that the first convolutions a r e quite sharp, the tubule keeping within the confines of the narrow portion of the lobule. The conrolutions, however, gradually increase in breadth, until at approximately one- Fig. 11 Composite drawing of several dissections to sliow the general arrangenicnt of anasto~nosingloops. The author has taken tlic liberty t o connect u p t h e broken or iiiifollowed ends of one dissection with those of another. R, branelling or f o r k ; 11, divcrticulunl; X, broken or unfollowed ends. third of the radial distance to the periphery they attain their maximum width. This inner portion of the contorted tubule, is directed straight toward the periphery, except for the sniall above-mentioned convolutions. Branchings a r e not frequent in this portion of the tubule but may occur at any point ; diverticula a r e relatively common. The outer portion of the contorted tubule shows a decreased tendency to reach the periphery a t once and frequently turns 196 FRANKLIN P. J O H S S O N baclrward upon itself in S-shaped loops. In this portion i t is not infrequent for a tubule t o leave i t s parent lobule aiid to enter a n adjacent lohule, occasioiially to return later. Once reachiiig the periphery the tubule ma>* r u n a short course along it aiid then clip toward tlie mediastiiiuin and after numerous convolutions, reach the surface again. Frequently N tubule will maltc several such contacts with the surface hcforc it fiiiallp contiiiucs its return course toward the rete testis t o join a straight tubule and complete a loop. F o r the most part the tubules of the testis form loops of this type, the two iiinci- ends lviiig in adjacent, o r nearadjaccnt lobules, a i d occasionally i n tlie same. Each such loop usuallp, though not invariably, possesses one o r more hixiiclics wliicli form adjacent loops, all united and in series with one another. To what cstent the loops are joined together in this maiinei- I am unable to state. Hubcr shows a sci*ics of scveii coniiectcd loops which he isolated from the rabloit 's testis. T feel confident that oiic coultl find iiincli larger series of loops in the human testis, inasmuch as loops without ii side branch (which when traced invariably formed iiiiother loop) are infrequent. Llii:istoniosing Im-aiiches mhicli comlilet e a continuous ring of contorted seminiferous tubule have actually been found aiitl vcrif id, hut i n my experience they arc infrequent, much iiiore 'io Ilian the original figure of S a p p y woulcl lead one to expect. Such rings map be present in either the inner or onter poi.tioiis of the tuhule a s sho~ v niii figure 1 3 , tuhulcs 10, 11 and 13. T n in:- dissections T liave f onnd lint two tubules \T-ith unqnestioiiablc 1)liiid c n d i i i g ~ . Tliese m i ' e both simple unbranched luhnles aiid in ncitlicr case reaclietl the loeripliei*p of the testis (figs.I and 10). One tubule showed in addition to its blind tcrmination four, and the other, fire blincllp ending diverticula. T n lmth cases the 1)lintl tcrmiiiations werc neatly coiled and hidden from view until uncovered hy careful dissection. S o hlintl terminatioiis havc heen found in the peripheral portion of the lobules. Should they esist, they are certainly not DISSECTIONS OF HUMAN SEMINIFEROUS TUBULES 197 numerous. Likewise, simple loops (without branches) have been found but they, too, a r e infrequent. Blindly terminating diverticula may be present in any portion of the contorted tubule but are seemingly more numerous and closer together in the inner portions. They are usually quite short, seldom exceeding a few millimeters in length. Their occurrence and distribution is variable, some tubules may be traced for great distances without encountering any, while in others they arc relativelv frequent. A definite variation has also been noted i n testes from different individuals. The longer diverticula at*efound in the outer portion of the tubules and it is undoubtedly these which Sappey has figured and described as blind ends of the tubules. It might at first appear to one that it is a matter of interpretation only whether such endings are to be regarded a s diverticula or a s the terminal ends of the tubules, in which interpretation the continuing portion of the tubule is to be regarded a s an wnastoniosing branch. Should the latter he the case we should find only one ending for each tubule instead of the numerous ‘diverticula’ scattered along its course. It seems a significant fact that these diverticula a r e of relatively late development as no mention is made of them by Bremer in his worl; on human fetal material and Huber failed to find them in the rabbit’s testis. Furthermore, Bremer shows that the loops develop as such from the early network of the medullary cords and it seems probable that all anastornosing branches and loops are derived from the network laid down before the tubules become patent. Ampulla-like swellings of the seminiferous tubules, such a s shown by Eberth, have been conspicuous by their absence in my preparations. The examination of a very large number of tubules taken from over a dozen different testes, resulted in the finding of only one. Frequently sharp kinks in the tubules give the appearance of ampulla-like swellings but when these a r e studied carefully from different angles they show no real enlargement of the tubule itself. 198 FRANKLIN P. J O H N S O N Very occasionally one finds a portion of a tuhulc ~ + i c h SIIOWS a nitirked d u c t ion in size. This nari*owing in thc clianicter of the tubule takes place very abruptly, the ~ ~ a r r o r n c d poi.tioii being approximately one-half the original cliameter or‘ the tubule. Such narrowed portions extend f o r distances of 1 to 10 em. (estimated) when they just as abrupt17 change to the usnnl dil-inicter of the tubule. Seemingly their convolutions are not a s great a s the larger tubulcs. That they are not clue to shrinkage causcd by the acid is suhstaiitiatecl by the fact that they i i i ~occasioiially seen in microscopic see t ions. An estimate of the total length of scniiniferous lubules, based upon a volume of 10 cc. of 13ai.eiichyna aiid a n average tubular diametcr of 0.2 imni. gives approxiiiiately 800 feet for cach testis. This figure is slig1itl;v lower tliiin niost other estimates which h a w been made. S o me;tsni*enients or estimates have been made upon the length of incliviclnal tubules or loops a s these vary so n i a i k d ~ ~ . ~ONCIlTTSIONS TT’hile the i*esnlts of the present investigalioii a r e in accord with those authors who state that the li~imanseminiferous tubule exists hotli in the form of complete 1001)s aiid blinc1l~oncling tubules, it is iiccessary to modifv this stntcmeut cis follosvs : By f a r the niost common tubules a r e the anastomosing loops, ~ d i i c l ai r c connectecl in a series of nndeterniined estent. Single, uiianastomosing loops have been found but a r e rare. Single unb~xncliiiigtubules a r e likewise pi-(.sent but a r c also rare; neither of the two found iii the course of this study reached the periphery of the testis. Complete rinqs of seminiferous tubules likewise were fouiid to be infrequent. Small divcrticula a r e frequently fonnd irregularly disti*ibutcd d o n g the course of the human seminiferous tubules. F r o m the standpoint of comparative anatomy it is of interest to note the following observations: I n certain marsupials the form of the tubules exist only in the form of DISSECTIONS O F H U M A N SEMINIFEROUS TUBULES 199 simple, unbranched and unanastomosing loops (DeBurlet and DeRuiter). I n the mouse simple loops are the most frequent f o r m ; doubled loops are occasionally found, and rarely it blindly ending tubule is ohserved (DeBurlet and DeRuiter). I n the rabbit Huber records only anastomosiiig loops; no blindly ending tubules. I n birds Huber describes a vast and complex network of tubules which resisted inr rave ling. I n the human fetus Brenier shows the presence of both loops and blind endings. I n the human adult simple loops are r a r e compound loops are numerous, while simple blind endinq tubules are very infrequent. Diverticula, so f a r as I am aware have been recorded only in the human seminiferous tubule. L I T E R A T U R E CITED DEBURLET,H. M. 1921 Zur Entmicklung und Morphologie des Saugerhodens. 11. Marsupialier. Zeitsclir. fur Anat. u. Entw., erste Abt., Bd. 61, S. 19-31. DEBURLET, H. M., AND H. J. DERUITER 1920 Zur Entwicklung und Morphologie des Saugcrhodens. I. Der IIodrn von blus musculus. Anat. Hcfte, Rd. 178, S. 32.5-383. BREMER,J. L. 1911 Morphology of the tubules of the human testis and epididymis. Am. J. Anat., vol. 17, pp. 393417. HURER,G. CARL 1916 A note on the morphology of the seminiferous tubules of birds. Anat. Rec., vol. 11, pp. 177-180. HUBER,G. CARL, AND G. It. CURl'IS 1918 Thc morphology of the SeminiferOUS tubules of niaminalia. Anat. Bee., vol. 7, pp. 207-220. RAPPEY, PH. C. 1874 Trait6 d'anatomie descriptive. Paris. ERERTH, I(. 1904 Die mannlichen Gesclilechtsorgane, i n Rardeleben 's IIandbuch der Anatomic des Menschen. Jena.