202 ISAO IIJIMA. 



being that the rachal end is bent forward less strongly than 

 the mediad end. 



The course of these fibres gradually changes towards the 

 termination of the ovary (fig. 1, t. oe.), in such a manner 

 that they finally assume a longitudinal direction, the ovary 

 itself dwindling to a point. The entire fibre, as above de- 

 scribed, is composed of two somewhat unsymmetrical halves, 

 which form together a complex curve, or a curve which faces 

 forward and inward at the same time. 



This system of muscle-fibres, as before stated, lies between 

 the second and the fourth stratum. 



The outlines of the cells which make up the fourth and 

 most of the second stratum are not represented in fig. 3, but 

 are more or less distinct in fig. 4. 



They do not have the angular form of closely-packed 

 cells, but are more or less rounded, and often separated by 

 considerable intervals. As is seen in fig. 4, the cells of the 

 second and the fourth stratum make up the greater part of 

 the ovary-wall. In most places the cells are two deep ; but 

 here and there a single cell appears to represent all that 

 can be seen of the two strata. The four strata just described 

 constitute what may be called the tunica propria of the 

 ovarial tubes. 



In regard to the fifth stratum, or the lining epithelium 

 (/. ep., fig. 4), I have found that it agrees perfectly with 

 what Leydig (No. 3, p. 549, mentioned in No. 15, p. 63) 

 has described in the case of Piscicola. This stratum is 

 composed of cells whic^i we may regard as epithelial, al- 

 though they do not present the form or arrangement 

 characteristic of most epithelial layers. Instead of forming 

 a continuous sheet of closely packed cells, they are so 

 loosely arranged that wide spaces are left between them, 

 except in the longitudinal area {ger.') borne by the rachis, 

 where they give rise to egg-strings. Outside this germinal 

 area the cells have an elongated spindle shape, the widest 

 part of which is filled by a nucleus "008 mm. in diameter. 

 Sometimes, especially in or near the germinal area, which 

 I shall call the germogen, cells with two nuclei are found. 

 These cells are usually so arranged that the longer axis 

 has a longitudinal direction. In section they appear as 

 small spherules attached to the wall of the ovary (/. ep., 

 fig. 4), with intercellular spaces measuring about "03 mm. 

 in width. 



As to the part played by the different muscle-fibres before 

 mentioned, it seems probable that the circular and semi- 

 circular fibres may be capable of producing peristaltic con- 



