SCALE INSECTS OF IMPORTANCE 347 



definite and well marked and of considerable size, the specimen is a third 

 stage, female A. perniciosus, though not yet gravid. The second 

 stage of these four species shows only the vague, somewhat parenthesis- 

 shaped thickenings mentioned above and seen in the second stage illustra- 

 tions. Of course the third stage of the other three species is marked by the 

 ventral glands even before the young appear within the body. In addi- 

 tion, the occasional difficulty in detecting the arrangement of the plates 

 must always be remembered. 



Key 



A Incisions wide and not very deep. Second pair of lobes small 

 when present. Median lobes rather broad. 

 B With fringing plates. Second pair of lobes rudimentary or want- 

 ing , A. ancylus 



BB Plates inconspicuous or wanting. Second pair of lobes usually 



distinct, though small A. ostreaeformis 



AA Incisions narrow. Lobes distinctly two pairs, of good size. 

 Median lobes rather narrow. 

 B Lobes nearly parallel. Fringing plates. Thickenings on either 

 side of the first incision subequal. No ventral glands even in 



gravid female A. perniciosus 



BB Lobes usually strongly approximating at tips. Plates incon- 

 spicuous or wanting. Thickenings on either side of first incisions 

 distinctly unequal. Ventral glands present in adult .A. f o r b e s i 



Aspidiotus forbesi Johns 



PLATES 12 AND 13, FIGURES I 



This species appears to approach A. perniciosus most closely in 

 general outline. (Compare each figure with the second figure of the same 

 plate, A. perniciosus.) It has four distinct lobes: the median ones 

 are notched on the outer margins and approximate at the tips, the second 

 ones are about half the size of the first and usually notched on the outer 

 margin. They also slant slightly inward, giving the species in general a 

 decidedly pigeon-toed appearance. The lobes are close together, because 

 the incisions (two pairs) are narrow. The first incisions are quite deep. 

 The second are not so much so. Prof. Johnson in the original description 

 and figure located four spines on each side of the median line ; these are 

 quite prominent. There is a fifth pair often visible near the union of the 

 terminal and the penultimate segments. In general the spines do not 



