24 ILLINOIS BIOLOGICAL MONOGRAPHS [144 



was estimated that from j&ve to ten larvae were injected into each host. 

 On account of the unfavorable conditions under which they were kept most 

 of the hosts died in the next few days. On the ninth, six of the infected 

 hosts were killed, the tissues teased out in salt solution and the sediment 

 examined. Two active larvae of Gordius robustus were found, both show- 

 ing signs of having begun their development. Both were lost in an attempt 

 to stain and mount them. On July 11 all hosts were dead except five. 

 When these were examined as before, four somewhat older, but still active 

 larvae were found. They also were lost by accident. 



I 2. Suspensions of larvae of Gordius robustus were injected into the 

 mouths of eleven young Locustidae on July 6. All died within a few days. 



I 4. Four young specimens of Orchelimum vulgare were infected as 

 in I 1 on July 9. On July 15 three of the specimens were examined and all 

 were found to be infected, yielding four small parasites (Figs. 9, 11). The 

 other host died. 



I 7. On July 11 fifty- three 'Locusiidd^.e mostly hy Orchelimum vulgare, 

 which had become easily recognizable by this time, were infected as before. 

 Specimens of this lot were killed for sectioning, at first every day and later 

 every two days. On July 15 an adult was examined but nothing found. 

 Another adult was killed and examined on July 22 and found to contain 

 three young worms, the longest being about 10mm. in length and the 

 smallest one about 5mm. On the following day the last three hosts were 

 examined and all were found to be infected, yielding 11 parasites ranging 

 from 3 to 10mm. in length. 



I 8. On July 14 seventy-eight Locustidae, mostly young Orchelimum 

 vulgare were infected. Some of these were again killed for sectioning and 

 staining as were those of I 7, One adult host was examined on July 22 and 

 found to be infected with one young parasite (Fig. 15). An examination of 

 ten more hosts on the following day yielded only a single parasite about 

 2mm. long in the coiled stage. On August 11 the last two specimens were 

 examined. Only one was infected and contained two parasites 25 and 

 30mm. long respectively. 



Later infection experiments proved less successful. Of 45 control hosts 

 not one was found to be infected. 



Of the specimens preserved for sectioning only seven from I 7 have 

 been thoroughly examined and six were found to be infected. Sections of 

 some of the specimens found are shown in figures 47-49 and 50-55. 



The results of these infection experiments show conclusively that an 

 intermediate host is not necessary for Gordius robustus. 



Infection in nature must begin in late June or early July and end the 

 latter part of July or in August. The young of Orchelimum vulgare collected 

 early in July were still small and could not have hatched more than a week 

 or so before they were collected. By the middle of September most of the 

 parasites are well along in their development. 



