HeproducUve Organs of Spams C'entrodoydns. 5 



dominates in the reproductive organ of a big Centrodontus, the other 

 sexual organ is so much reduced as hardly to he recognised ; it seems, 

 indeed, to be exhausted. This condition is seen in Fig. 40 (natural 

 size), where, while the ovar}^ is large, the testis is reduced to a thread. 

 In the previous case the fishes were, I think, preparing to spawn for 

 the first time. In the latter case the ovar^' had been ripe before. 



The conclusion, then, is that the organ when it first ripens is 

 hermaphrodite ; at the next spawning it may be unisexual. I do not 

 know whether the hermaphrodite condition ma}'- persist in the second 

 spawning. It seems not improbable that where in the hermaphrodite 

 condition one sex specially predominates, that will be the sole function- 

 ing sex at the next spawning. 



Organs in which two sexes were equally represented are given in 

 Figs. 17 and 21. The testis was white, ripe. It was possible to make 

 out the heads of the sperms, but not their tails. The ovary had an 

 amber tinge. In one specimen, Fig. 21, the eggs measured up to 

 •1mm. in diameter, while in the other they reached a diameter of 

 •15mm. A section of the latter reproductive organ is given in Fig. 18. 



Where the male predominated the testes were nearly all large and 

 white. In one case the testis was ripe. It was pastj^ in consistency ; 

 the heads of the sperms, but not their tails, were made out. The ovarj' 

 was very small and colourless ; it «as enclosed in a deep groove in the 

 edge of the testis. The ovary was 4nnn. broad, and it was sunk flush, 

 almost out of sight. In Fig. 3a the groove has been opened out to 

 expose the ovary (ov.). It is quite a normal immature ovary, with 

 ridges filled with minute eggs, measuring about "OSmm. in diameter. 

 In another fish the ovary was only partly sunk in the groove (Fig. 9). 

 The eggs measured from -OSmm.-^Oymm. in diameter. They were 

 colourless, non-yolked, and clear. These ova have a distinct round 

 nucleus. 



Sometimes the ova •05mni. in diameter, although translucent, are 

 not perfectly clear, as the substance of the egg is granular ; they have 

 an exhausted appearance. 



The dorsal mesentery {dm., Fig. 38), which supports the organ to 

 the roof of the abdominal cavity, runs along the ovary. The testis 

 forms the free ventral border of the united organ. Sometimes the 

 side of the ovary is attached to the testis ; in other cases the attach- 

 ment is very slight (e.j/.. Fig. 25). 



A section across a predominating male organ, at a quarter of its 

 length from its posterior extremity, is shown in Fig. 8. The ovary is 

 a sac, on the inside wall of which are arranged ridges filled with ova. 

 The ridges run longitudinally. In some cases, e.g., in a fish 43*5cm. 

 long, the egg-ridges rose little, if any, above the surface of the ovary. 

 The eggs were clear and nucleated, the largest measuring "OGmm. in 

 diameter. Part of the inside wall of the ovary is without ova, i.e., the 

 oviducal part. 



The ovary and the testis do not communicate. Between the two 

 there is a layer of lacunar, or spongy honeycomb tissue, which extends 

 into the wall of the ovary (sp., Fig. 8). This lacunar tissue acts as the 

 vas deferens. 



The testis is of an honeycomb structure. A surface view of a portion 

 of the ripe testis is shown in Fig. 16. It is seen from the outside. 

 The spaces are occupied with sperms. The internal structure of a 

 whitish developing testis consists essentiall)' of tubules (cf., Fig. 11). 



