56 
Female (boiled in potash) turns the liquid a pale, port-wine color. Derm 
colorless, with numerous gland spots and some small spines. Antennae 
and legs pale ochreous, comparatively large. Antennae distinctly 
9-jointed: 3 longest, 2 nearly as long and decidedly stouter ; 1, 4, 5, (i, 
and 9 subequal; 7 and 8 subequal and shortest: formula 3, 2 (1, t, 6, 
6,9) (7, 8); 1 with two stout hairs near its end, 2 with a long hair, 8 
with a pair of hairs near the end; remaining joints each with a 
whorl of hairs; last joint with also apical hairs representing a second 
whorl. Legs ordinary: coxa very large, with a whorl of bristles near 
its end: trochanter and femur with scattered bristles; femur with an 
erect hair on its inner face, just before its middle. Tarsus less than 
half as long as tibia; tibia with about five bristles on its inner face and 
six on outer. Tarsus with bristles. ( law long, not much curved : digi- 
tizes of claw of fair size, expanding rather gradually to their bulbous 
ends. Tarsal digit ales wanting. 
Habitat, — Japan, precise locality unknown, on " Gumi." (Takahashi; 
Dept. Agr. No. 5942.) The scales occur on the undersides of the leaves, 
along the midrib. What kt (iumi " is, I do not know, but it has entire 
rather hairy leaves about 40 mm. long, suggestive of some solanaceous 
or serophulariaceous plant 
At lirst sight the species looks like a very much developed PtiJri- 
naria camellicola, but the texture of the ovisac suggests Lichtensia. I 
had actually described it as a new Lichtensia and had sent the MS. 
to Washington, when Mr. Pergande, having occasion to examine the 
insect, discovered the extraordinary error into which I had fallen. The 
specimens were much attacked by parasites (a species of Comys, 1 learn 
from 31 r. Howard), and the legs, antenna', etc., were detached. Thus, 
having gotten the erroneous idea that the thing was a lecaniid, I de- 
scribed from what 1 could see, notwithstanding the absence of the ano- 
genital parts, etc. On receiving Mr. Pergande's statement, I boiled down 
a new specimen, and was fortunate enough to see the anal ring, perfectly 
normal for Phenacoccus, to which the insect unquestionably belongs. L 
mention these incidents because such errors are always interesting, 
throwing light on the probability of error in scientific writings. 1 have 
sometimes seen it stated that so-and-so could not have made a certain 
mistake, because he knew better; but a careful analysis of mistakes 
will show that a large percentage would not have been made if the 
writer had known less. For example, a traveler in a foreign country 
will often announce that he saw some bird or insect very familiar to 
him at home, and when it is denied that the species occurs there he will 
indignantly ask whether we suppose he does not know the common so- 
and-so. As a matter of fact, he has been misled by a superficial resem- 
blance; whereas had the object been quite unfamiliar to him he would 
have taken pains to arrive at its correct identification, probably with 
success. 
