4 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 67 



The genus AmphoropJwra seems to be world-wide in its distribution. 

 From data at hand various species of the genus are found in North 

 America, South America, Europe, Africa, India, and Japan. 



Certain groups of plants are favored as hosts. The genus Ruhus 

 has eight of the species discussed in this paper. Rihes has four 

 species. In the family Ericaceae the genus Vaccinium has one, 

 Gaultheria one, Epigoea one, Azalea one, and Rhododendron two spe- 

 cies. The family Compositae is host to five species. This family 

 is known to be the summer host of one of these five and will probably 

 prove to be the same for the others. The Coniferae furnish hosts 

 for one species in this country and one in England. 



There are practically no data on the alternation of hosts in this 

 genus. Only one, and probably two species are known to have alter- 

 nate hosts. Some of the others seem to live on the same plants 

 throughout the year. 



Very little can be said on the origin and phylogeny of the group, 

 as the species are too imperfectly known at present. I am not 

 in a position to say where the genus originated or on what plant. 

 In fact, I can not be absolutely certain that I am dealing with a 

 phyllogenetic group. It seems probable that those species on a cer- 

 tain hosf group originated from a common ancestor, as they seem to 

 be rather closely related. 



In the group of forms mentioned as living on Compositae there are 

 several very closely related ones. These include cosmopolitana 

 {lactucae Kaltenbach), carduellinum Theobald, oleraceae v. d. Goot, 

 formosana Takahashi, sonchifoliae Takahashi. There are also some 

 other specimens in the National Collection which could be described 

 as new species with about as much difference as is given for most of the 

 species mentioned, but I hesitate to do this until more is known of 

 them. This group is so closely related that it is often difficult to 

 place an individual specimen. There should be extensive rearings 

 to settle definitely the number of species concerned. Until this is 

 done I am retaining as valid species those which have been described, 

 with the exception of sonchifoliae Takahashi. He admits that it is 

 very near oleraceae v. d. Goot, but his differences do not hold when 

 checked against metatype specimens. 



Certain species which have been placed in this genus by various 

 writers are not included here, since they belong to the group men- 

 tioned as having imbricated antennal tubercles. These include 

 rhinanthi Schouteden, subterrans Wilson, magnoliae Essig and 

 Kuwana, lonicericola Takahashi, and hydrangeae Matsumura. 



Other species properly belonging to Megoura have been placed in 

 Amphorophora at times and there is some evidence for considering 

 Megoura as a synonym of Amphorophora. However, I have decided 

 to retain it as a valid genus, separating the two on the shape of the 



