179] LIFE HISTORY OP GORDIUS AND PARAGORDIUS—MA Y 59 



indicated by the figure of the first specimen described by Wesenberg-Lund. 

 In neither case can one speak of a true protection of eggs or young by the 

 parents. Indeed, in case of the specimens observed by Wesenberg-Lund in 

 the close masses it is very likely that most of the eggs had dropped to the 

 bottom of the pond long before the larvae were ready to hatch. 



Behavior of larvae 

 Little is known to the present time in regard to the habits of the larvae, 

 except that they penetrate the tissues of a great many animals and in most 

 of them become encysted and perish later. Cort (1915) even found such 

 encysted larvae in trematodes. Villot (1891) and Camerano (1897) con- 

 clude from their own observations that such encysted forms are invar- 

 iably lost and can not undergo further development. My observations on 

 Gordius rohustus show that an encysted stage in that species is not necessary 

 and also that larvae that have lived free in the water for some time are 

 usually incapable of development. 



Infection and intermediate host 

 It is possible in the light of the present evidence to show that some of the 

 former theories of infection are not tenable. The most commonly accepted 

 theory in Europe has been that which assumes an active migration of the 

 Gordius larvae into the larvae of aquatic insects or into other soft bodied 

 aquatic animals and a consequent passive migration into a second host, 

 usually an aquatic insect, which devours the first host. This theory finds 

 its support again in the recent preliminary account of the life history of 

 Gordius tolosanus published by Hans Blunck. He differs from the older 

 views in that they assume that the adult Dytiscus, to mention a specific 

 case, devoured the infected first host while he claims that the larval Dytis- 

 cus ingests the first host and becomes infected. The older theories seemed 

 very logical in cases where the final host was an aquatic carnivore, but 

 were difficult to apply where it was a supposed herbivore like a grasshopper 

 or a cricket. Montgomery supposed in such cases that the first hosts were 

 perhaps Mayfly larvae, that the encysted parasites were carried out of the 

 water by the emergence of the Mayflies, and that they were liberated at 

 the death of the insect, remained for a time on grass or leaves, and were 

 taken into the final host with the vegetable food. Even as late as 1904 he 

 had not discovered that Gryllus abreviatus (assimilis), which he knew to 

 be the host of Paragordius varius is not only an omnivore but a cannibal 

 and that it is fond of its nightly bath. Assuming that the host is truly 

 terrestrial, he went so far as to undertake experiments on desiccation of 

 worms that had just emerged and to formulate theories in regard to the 

 chances a worm deposited on dry land had for getting back to the water. 

 Observations made during the present investigations show that all hosts 

 of both species of Gordiacea here considered are neither truly terrestrial 



