1913.] NATURAL SCIENCES OF PHILADELPHIA. 139 



The first (direct) mode seems impossible for the following reasons : 



A. In several cases we found infested birds whose cagemates w^ere 

 free. 



B. Several pigeons were fed on eggs or actually had them placed in 

 the crop by rubber tube and syringe. Killed at intervals of a month, 

 proventricle and gizzard showed neither micro- nor macroscopic 

 involvement. The last bird killed had been treated three months 

 previously. Pigeons ought to be susceptible to these worms. Our 

 records show that a Barbary turtle dove and white-crowned pigeon 

 died with them in September, 1911, and three blue-crowned pigeons 

 were similarly infested in September, 1907. 



C. In order to get a perfect blood relationship, ova and embryos 

 were placed in the crop of newly arrived parrots which had been 

 tested by the improved method and passed. Another parrot was 

 kept in a cage soiled by a verminous bird. One bird ched in four 

 weeks ^vith pneumonia. It showed no worms. The others were 

 examined regularly and so far have shown no ova in the stools. 



D. Roseate cockatoo (No. 166) was placed in a cage with infested 

 roseate cockatoo (No. 120) and drinking water allowed to become 

 soiled on August 28, 1912. The birds were separated twelve weeks 

 later and the healthy cockatoo found still healthy, as shown by 

 examination of droppings. 



In each of the four conditions cited above an opportunity was 

 afforded for the ova to develop by the first mode. They failed 

 to do so. 



In testing out the second mode, embryo worms were fed to pigeons 

 and parrots in the same way that eggs were administered in the 

 preceding experiments. The pigeons were killed at one month 

 intervals and the proventricle and gizzard examined by the micro- 

 scope. No infestation was found. A roseate cockatoo also received 

 the larva. In twelve months no ova have been found in the droppings 

 upon repeated examinations. The remaining possibility — entrance 

 by skin penetration — is now being tried. 



The third mode which would involve an intermediate non-biting 

 host, does not harmonize with our ideas of nematode transmission. 

 In only one case, that of Spiroptera sanguinolenta of the dog, could I 

 find such an example. The cockroach is the secondary host here, 

 and as our aviary harbored many such insects, attention was directed 

 to them. They were fed on cornstarch into which spiroptera ova 

 were mixed. The roaches were dissected at intervals of two days, 

 and though the eggs were found in the cloacal contents, no develop- 



