190 BULLETIN OF THE BUREAU OF FISHERIES. 



(sm) filling the posterior half of the testis, followed by a zone of sperm daughter cells 

 (sa), and these transforming into the sperms (se) in the anterior end of the testis and 

 filling the lumen of the sperm duct. In a cross section through the third thorax segment 

 (fig. 67, pi. xm) the posterior testis is seen to be saddled upon the dorsal surface of the 

 intestine, while the sperm ducts appear on either side opposite the center of the testis. 

 Beneath the intestine and close to the ventral surface is the broad thick trunk of the 

 inf ra-oesophageal ganglion. 



In the female the ovaries are also spindle-shaped and flattened dorsoventrally. 

 The posterior end is attached to the dorsal body wall, and from the anterior end the 

 oviduct leads back to the vulvae in the genital segment. The posterior portion of the 

 oviduct has not yet become glandular but is like the anterior portion, and the whole 

 duct is without convolutions. The spermatophores of the male are attached to the 

 vulvae and their contents are discharged into the oviducts. The sperms probably 

 remain here during the subsequent growth of the female's body and the ripening of the 

 eggs, and when the latter pass down the oviduct they are fertilized just before reaching 

 the glandular portion where the cement substance forms the egg membranes. 



There is no frontal gland in these larvae, the space in the anterior portion of the 

 head being occupied by the powerful muscles which operate the mouth parts. 



ADDITIONAL DATA. 



Fifty gars whose gUls were infested with the copepodid larvae of LerncBa variabilis 

 were obtained from Patterson Lake, near Fairport, Iowa, August 15, 1915, and were 

 placed in cages in one of the fish ponds. 



Three or four of these were examined twice a week in order to determine if possible 

 how long the larvae remained and what became of the two sexes after fertilization. 

 The author could only carry the examination up to September i, but W. B. Gorham, 

 one of the regular station staff, very kindly continued it through the month of Septem- 

 ber and up to October 4, sending the specimens found upon the gills to the author for 

 examination. H. W. Clark also collected samples of plankton, which were sent simi- 

 larly for examination. In this way it has been determined that : 



1. These larvae appear first about the latter part of June and may be found upon 

 the gills of the short-nosed gar until the middle of October. 



2. Each copepodid stage occupies from 10 to 14 days and apparently is accom- 

 panied by several molts. 



3. Fertilization occurs during the fourth copepodid stage, as was proven by finding 

 the two sexes fastened together upon the gills and by witnessing the union once or 

 twice while handling the larvse in watch glasses. 



4. After fertilization the female apparently seeks out her final host at once, since 

 nothing more is seen of her either on the gills or in the free-swimming plankton. The 

 male, on the contrary, remains upon the gills of the temporary host, and specimens 

 were found every time the gills were examined. 



5. During the several copepodid stages and for some time subsequently the two 

 sexes are found in about equal numbers. Then the proportion of females diminishes 

 steadily and by the last of September only an occasional female is left, while the males 

 are apparently undiminished. 



6. Careful and repeated search of the plankton does not reveal any of these larvae, 

 and hence we can only conclude that the male does not become a free-swimmer for a 

 second time, but remains upon the gills of the first host until his death. 



