During or after the completion of the above changes a number of bodies usually spoken of as test-cells make their appearance in the superficial protoplasm of the egg, which by the time the egg is ripe arrange themselves in many species as a definite layer round the periphery of the ovum. These bodies have received their name from the opinion, now known to be erroneous (Hertwig and Semper), that they eventually migrated into the test or mantle of the embryo which becomes developed round the ovum. By Kowalevsky (No. [58]) these bodies are regarded as true cells, and are believed to be formed by some of the cells of the original follicular epithelium making their way into the vitellus of the ovum and multiplying there. By Kupffer (No. [60]), and Giard (No. [61]), and Fol, they are also regarded as true cells but are believed to originate spontaneously in the vitellus. Finally by Semper they are believed not to be cells, but to be amœboid protoplasmic bodies which are pressed out from the vitellus under the stimulus of the sea-water or otherwise.
They do not according to this author naturally appear till the ovum is quite ripe, though they can be artificially produced at an earlier period by the action of reagents or sea-water. When produced in the natural course of things the vitellus undergoes a contraction. They are without any apparent function, and play no part in the embryonic development. Semper’s results are very peculiar, but owing to the careful study which his paper displays they no doubt deserve attention. Further investigations are however very desirable. Kowalevsky from his researches on Pyrosoma (No. [59]) adheres to his first opinion, though he abandons the view that these cells are connected with the formation of the test.
In the passage of the egg through the oviduct the vacuolated follicle cells grow out into very peculiar long processes or villi. In Ascidia canina these processes become as long as the whole diameter of the vitellus (Kupffer, No. [60]).
In Amphioxus and the Craniata the ova are developed as in the Chætopoda, Gephyrea, etc., from specialized germinal cells of the peritoneal epithelium.
In Amphioxus the germinal epithelium which constitutes the essential part of the ovary is divided into a number of distinct segments: in the Craniata no such division is observable.
In young examples of Amphioxus the generative organs are in an indifferent condition, and the two sexes cannot be distinguished. They form isolated horse-shoe shaped masses of cells, which occupy a position at the base of the myotomes, in the intervals between the successive segments; and extend from the hinder end of the respiratory sack to the abdominal pore. They are situated in the proper body cavity, and are surrounded by the peritoneal membrane. Each generative mass is at first solid, and is formed of an outer layer of more flattened cells and an inner mass of large rounded or polygonal cells. In its interior there appears at a somewhat later period a central cavity. After the cavity has appeared the sexes can be distinguished by the different behaviour of the cells.
In all the Craniata, the ovary forms a paired ridge (unless single by abortion or fusion) attached by a mesentery to the dorsal wall of a more or less extended region of the abdominal cavity. This ridge is at first identical in the two sexes, and arises at an early period of embryonic life. It is essentially formed of a thickening of the peritoneal epithelium, and in Osseous Fish, Ganoids (?) and Amphibia the ovary remains during embryonic life nearly in this condition, though a small prominence of the adjacent stroma also becomes formed. In other Craniata the ridge, though at first in this condition, very soon becomes much more prominent, and is formed of a central core of stroma enclosed in the germinal epithelium ([fig. 18]).
Fig. 18. Transverse section through the ovary of a young embryo of Scyllium canicula, to shew the primitive germinal cells (po) lying in the germinal epithelium on the outer side of the ovarian ridge.