ON THE DEVELOPMENT OF DERMATOBIA HOMINIS. 509 



beneath the skin, whether they change in form,— these earlier details are entirely 

 unknown, but within about two weeks after the infection the larva has in the 

 human host arrived at the stage designated by Blanchard as ver macaque; the date 

 is positive in the case reported by Matas ('87), in which I have studied the larvae. 

 The host was stung by a fly June 11 and the larva; removed on June 27. At the end 

 of about five weeks the larva molts or has just passed that epoch, as in the case of 

 the two larvae from man described in this paper. Data are lacking in most cases for 

 a more precise determination of these limits and no doubt external factors modify 

 them somewhat at least, but no larvae of the ver macaque stage have been obtained 

 which were known to be older than this and none of the torcel stage which were 

 definitely younger. 



The fully developed specimens of the torcel stage are more than two months old, 

 and it may be even that another molt intervenes between the older specimen that 

 I have described and the form from the same country illustrated by Grube and con- 

 strued as the mature form of the torcel stage. Numerous minor details which can- 

 not be profitably enumerated hei;e favor this hypothesis, but it must be established 

 by evidence of another kind. According to Coquerel et Salle ('59) the larva 

 remains ordinarily three months in the flesh; at the end of that time it drops to the 

 ground and transforms. This was also the age of the larvae described by Magalhaes 

 (Blanchard, '92, pp. 144-145), which he thought ready to pupate. The pupal period 

 has been determined by Goudot ('45), who collected larvae from the ground where 

 infested cattle had passed the night and raised the perfect insect. In Colombia this 

 required in the single instance about six weeks. 



No doubt the life cycle is periodic, and yet there is not sufficient positive evidence 

 to warrant drawing conclusions. It is suggestive that the cases of human parasitism 

 bear dates, so far as I have been able to find any at all in the accounts at hand, which 

 fall in the months of April to July, inclusive. 



2. H est — The most diverse animals are at times the host of the species under 

 consideration. Commonly the larva parasitizes in the skin of cattle, pigs, and dogs; 

 it occurs less frequently in man and rarely in the mule, and writers have commented 

 upon its absence from the horse. It also occurs in the agouti (Bonnet), jaguar (Roulin), 

 in various monkeys as the guariba, or Mycetes ursinus (Bates), and a species of Cebus 

 (Carriker). I have also to record the observations of Carriker on its occurrence in 

 the toucan (Rhamphastos tocard) and in an ant-thrush (Formicarius sp.). At first 

 I thought that this was the initial record of its presence in birds, but this appears to 

 be incorrect. Guy on ('36), whose paper is known to me only by the citations given 

 by Coquerel et Salle" ('62), refers to the fact that these larvae occur in birds as well as 



