460 REPORT OF COMMISSIONER OF FISH AND FISHERIES. [6] 



mackerel, which, as in the latter, no doubt causes the ovum to float 

 <lunu^ (levoloimient. 



With the further growth and maturation of the ovicells in the lobules 

 of the iunctionally active ovarj^ it becomes possible, after they have 

 attained the dimensions of about .3™™, to discern that each one is in- 

 closed in a more or less well-defined covering of small cells; the ovisac, 

 ovarian capsule or follicle, in its early stages at least, is often, if not usu- 

 ally, composed of flattened or of columnar cells. The history of the de- 

 velopment of the ovary in bony fishes is not well known, nnd little has 

 been written upon the subject of a satisfactory character, except more 

 recently by McLeod, Waldeyer, and Brock, from whose researches it 

 would appear that the ovary makes its appearance as a differentiation 

 of two bands of peritoneal epithelium placed on the dorsal side of the 

 body cavity, on either side of the mesentery. So far as known, the early 

 development of the generative tissues is similar in sharks and true 

 fishes. The bands of primitive germinal cells are known as the germi- 

 nal epithelium. The reproductive cells are distinguished from the ad- 

 jacent indifierent epithelium at a very early stage, and are known as 

 jprimitive ova. These ijrimitive ova or germinal cells become either ova 

 or spermatozoa, it being impossible to distinguish what will be their fate 

 when they first make their api^earance. In some forms the primitive 

 ova are soon aggregated into masses, which break up into ampullje, 

 which are afterwards attached to tubes derived from the smaller invest- 

 ing and indifferent cells. In most Teleosts the ovary in a developed state 

 is lobulated, each lobule consisting of great numbers of ova in different 

 •stages of development. These lobules may be arranged in a longitudi- 

 nal or transverse manner; the latter appears to be the most usual mode. 

 It is doubtless true that in some fishes the ovarian lobules, when trans- 

 versely arranged, correspond more or less closely to the muscular and 

 vertebral segments. In other cases no such arrangement is apparent; 

 the germinal fold may be rolled upon itself longitudinally, or the ova- 

 rian rudiment may ai)pear to be derived from the anterior portion of the 

 dorsal peritoneal wall of the abdominal cavity, as in Gamhusia patruelis^ 

 since in embryos of this species which have not yet absorbed their yelk- 

 sacks we may see the primitive generative structures as a pair of cylin- 

 <lrical organs lying in the upper part of the abdominal cavity, attached 

 to the peritoneum only at their anterior ends. Later, when the ovary 

 is developed, no trace of lobulation is apparent, and the small mnnber 

 of ova which are matured remind one of a bunch of grapes attached to 

 the stem, the latter representing ideally the vessels which nourish the 

 growing ova. In other forms the arrangement is very diiierent. In 

 the cod the ovary is enormously' developed, and is an internally lobu- 

 lated and closed, paired organ, opening outward by way of a wide duct 

 behind the vent ; the body of the ovary itself extends some distance 

 behind the vent into a prolongation of the abdominal cavity, where its 

 two halves are conjoined. In very immature stages of development of 



