126 THE COCCIDAE 



to make different species of tliem. While Smith has shown of how 

 little value most of the comparative characters used in separating 

 species are, he has called attention to the use of several new struc- 

 tural characters, as the cephalabiae and caudalabiae which were 

 first figured by Berlese, the cerari and the number of conical setae 

 contained in each which had been previously used by Marchal 

 unbeknown to Smith, the cisanal and obanal setae, the cerores of 

 the anal ring and their arrangement in rows, the difference in the 

 form and extent of the t\vo rows, and the dentacerores of the mesal 

 row. It has also been noted b}' Matheson that in the ease of 

 Pseudococcus citri that there is a great variation in the size of 

 the adult females when they commence laying their eggs, as he 

 states, "often what I, judging from the size only, considered were 

 nymphs in the third stage would prove to be mature females and 

 commence egg laying." This is, undoubtedly true of many if not 

 most other species of coeeids. 



The subfamily Eriococcinae includes two groups of genera 

 which are easily distinguished in the first nymphal stage. In one 

 group the dorsum bears rows of blunt conical setae and in the 

 other the dorsum bears fine pointed setae. This latter group 

 clearly includes the generalized genera and has been designated 

 as a tribe by Cockerell under the name Dactylopiini. The group 

 with the rows of blunt dorsal setae is clearly the specialized end 

 of the series, genera in which the ordinary pointed setae have 

 become changed in form and acquired a symmetrical arrangement 

 and have been designated as a tribe by the same author under the 

 name of Ericoccini. Whether these two groups of genera repre- 

 sent distinct subfamilies as the form and arrangement of the setae 

 of the body might suggest, is not clear. Our knowledge of the 

 anatomy of the adult female is very imperfect. What is needed 

 in this subfamily is not the description of more new species and 

 genera, but a careful study of the anatomj' of the described forms 

 and an extended search for additional characters upon wliich their 

 classification and their phylogeny can be based. 



The following table to the genera is based in great part upon 

 descriptions and is faulty not only in construction but undoubtedly 

 in the characters used. Where a difference in number of antennal 

 segments is shown, specimens should be traced through both sides 

 of the table. I have egg laying females of what I believe are 

 undoubtedly species of Phenacoccus with less than nine segments 



