Characters in Paratrombium — Newell 
County, Washington. Sandy mud, in grass. 
July 27, 1954. Collected by the writer. At the 
point where this species was found, Sherwood 
Creek is estuarine, opening into Case Inlet of 
Puget Sound. The mites were collected on a 
low mound which was nearly covered by 
high tide. 
remarks: The correlation between the 
larva and adult described above was estab- 
lished by rearing from eggs laid by the 
female collected at the type locality. 
The relationship of this species to de- 
scribed forms is somewhat obscure, but there 
is no doubt that it is congeneric with P. 
quadriseta n. sp. The adult keys out with some 
slight inconsistencies to the genus Caeno- 
thromhium Oudemans, 1927, and agrees with 
the type of that genus in a number of details, 
including the general form of the crista, the 
double, stalked eyes and the absence of a 
distinct ctenidium on the tibia of the palp 
(there is definitely no distinct ctenidium in 
P. b. and apparently none in C. caloris). How- 
ever, the larva does not appear to be conge- 
neric with the larva of C. miniatum Womersley, 
1934, for which species Womersley has estab- 
lished a correlation. Differences, as far as can 
be judged from Womersley’s figures of this 
larva (1939^.' 156), are in the chaetotaxy of 
the scutum, the absence of the heavy comb- 
like seta of coxa I and the undivided (?) 
odontus. The drawings in the original de- 
scription of the larva are not complete, but 
they probably are fairly accurate in these three 
respects. The larva of P. bidactylus keys out to 
the third dichotomy in the key given by Thor 
and Willmann (1947: 484). At this point fur- 
ther separation was made on the basis of the 
claws of tarsus III— whether these are normal, 
or whether the posterior claw (inner-claw) is 
deformed. Neither of these fits P. bidactylus or 
P. quadriseta accurately, and the key, as well 
as the diagnosis of the genus given by Thor 
and Willmann, were in error on this point. 
The fourth dichotomy of that key is totally 
unreliable since it is based upon the presence 
or absence of setigerous sclerites behind the 
365 
postscutum. It is quite likely that these have 
been overlooked by many authors in describ- 
ing their species because of refractive index 
difficulties. 
An examination of the drawings of the 
larva of P. egregium presented by Bruyant, 
Oudemans, and Thor and Willmann shows a 
close similarity between P. bidactylus and P. 
egregium Bruyant, 1910. The similarity is indi- 
cated in the shape of the scutum, the form of 
the prosensillar setae and their insertion at 
the ends of long canals, the heavy pectinate 
tritorostral setae, the comblike seta of coxa I 
and the chaetotaxy of the body, especially the 
ventral setae. These forms are extremely simi- 
lar, and the resemblance is far too close to be 
a consequence of convergence; hence we can 
assume that they are actually very closely 
related. 
The foregoing discussion leaves a number 
of important questions. For one thing, is the 
genus Caenothrombium made up of genetically 
related species, or is it a composite? Was 
Womersley correct in assigning his Caeno- 
thrombium miniatum to that genus? Was the 
female which laid the eggs from which 
Womersley described the larva actually C. 
miniatum or a different species? There seems 
to be no way in which these questions can be 
resolved by reference to the literature, so 
further speculation about them is pointless. 
Paratrombium quadriseta, new species 
female: Idiosoma, 1,768 /jl long to tip of 
scutum, 1,092 11 wide, length/width 1.62. 
Scutum of dissected female (Fig. 50) resem- 
bling that of P. bidactylus in most details. 
However, the anterolateral lobes of the scu- 
tum, extending from about .25 to the anterior 
limits of the scutum bear only 13 to 14 setae 
on each side (P. b.: 22 to 25 setae on each 
side). In the present species there is a wide 
interval between the three anteromedian setae 
and those more laterally placed, whereas in 
P. bidactylus , this wide gap was not found. 
Width of scutum equal to .67 of the median 
length. Sensilla situated at .40, appearing 
