44 ARTHUR ROBINSON. 



young at a birth, and bring them forth at frequent intervals. Again, 

 the pouch is well developed in the elephant, an animal that gives 

 birth to few offspring at long intervals. Indeed, the sac is present 

 in a more or less well developed form throughout all the mammalian 

 orders, and all the modifications of it, from the simplest to the most 

 complete, are presented by one order, namely, the order of rodents. 



It is interesting to note that this shutting off of a portion of the 

 body cavity to form an ovarian sac is neither confined to mammals 

 nor to vertebrates, but is also found in the invertebrate kingdom. 



In the Chsetopods the ovary is placed in the body cavity, and the 

 ova are shed from its surface into that cavity, and escape from 

 it by an oviduct. In leeches the ovary is enclosed in a sac which 

 communicates with the exterior by an oviduct, but has no opening 

 into the body cavity. If, as Balfour states,* the affinities of the 

 leech are with the Chsetopoda, this ovary must have become shut off 

 from the body cavity in a secondary manner ; and Bournef supposes 

 the shutting off to have occurred at some time prior to the appearance 

 of haemoglobin in the blood fluid, for he finds the ovarian sac filled 

 with a fluid similar to the blood, except that it contains no haemo- 

 globin. 



Again, in some fishes the ova are merely cast off from the ovaries 

 into the body cavity, from which they escape by abdominal pores 

 (Marsipobranchi). In others an oviduct is present, the abdominal 

 orifice of which lies close to the ovary (Elasmobranchii and Ganoidei). 

 In many Teleostei the ovary is placed in a sac which lies in the 

 abdominal cavity, and is attached to the abdominal wall by a 

 mesovarium. The lower portion of the sac is lined by a different 

 epithelium, and forms a duct leading to the exterior of the body 

 It seems very probable, taking into consideration what occurs in 

 mammals and leeches, that the sac-like teleostean ovary is a secondary 

 formation, and is developed in some way from those cases in which 

 there is an ovary and a Fallopian tube. 



There appears, however, to be no evidence that such is the case 

 and the development of the sac-like ovary has not, so far as I have 

 been able to ascertain, been described. 



* Comp. Embryol., vol. i., p. 35. 



t Quar. Jour. Micro. Soc, No. xlv., 473. 



