SUBFAMILY KERMESIINAE 193 



subfamily have been defined in all the diagnoses of the subfamilies 

 as body naked and gall-like in form. It is not only unwise but 

 improper to designate the body of any of the naked coccids as the 

 scale as is frequently done by certain writers. The body in some 

 species is slightly longitudinally or transversely constricted at 

 middle or furrowed, so that it is more or less reniform. A few 

 species resemble certain of the naked lecaniids, others are elongate 

 and more swollen. The species also vary in that some occur singly 

 or at most in groups of three or four upon a branch while others 

 occur in such numbers as to completely cover a considerable por- 

 tion of the stem. 



The appendages are present on the ventral aspect of the adult 

 female. The rostrum consists of two segments. The antennae and 

 legs of the adult female may be well developed. The former con- 

 sisting of six or less segments, or they may be rudimentary or 

 wanting. The thoracic spiracles are present, normal in position, 

 and surrounded by numerous spiracerores. The anal lobes and the 

 anal setae are wanting as well as the anal ring and the anal ring 

 setae. The ventral aspect of the body may bear numerous cerores 

 from which is excreted the white wax located between the body of 

 the female and the host-plant, but this wax is generally not ex- 

 posed until the insect is removed. The exposed surface of the 

 body may bear cerores from which is excreted the powdery wax 

 which is spread over its surface. Some species bear numerous 

 short ceratubae. The genus Kermes was originally associated with 

 the lecaniids. In order to justify this location, it was necessary 

 to show the presence of an anal cleft and opercula. Some writers 

 have indicated the presence of such structures, but these structures 

 are wanting in most species and it is very doubtful if they or any 

 other structures similar to them occur in any species. Maskell in 

 defining his group Hemmicoccidinae, which is the equivalent of 

 the Kermesiinae, makes the following statements : ' ' Adult female 

 exhibiting the anal cleft and the lobes of the Lecanidae ; naked or 

 covered. Larvae presenting at the extremity of the abdomen the 

 anal tubercles of the Coccidae. From the foregoing characters the 

 group is very evidently intermediate between the Lecanids and 

 Coccids." 



While the adult females of certain species of Kermes do show 

 a superficial resemblance to the naked lecaniids, the first stage 

 nymphs show clearly their true relationship. They are distinctly 

 eriococcid, not only in general appearance, but in structure. The 



