Taylor (1915, p. 96) described Lupcrosanriis compresicorpus on the basis of a single female specimen from Batan Province, Luzon Island in the Philippines, noting at the time that he was not sure of its generic position. Later (1922, p. 103), he made this species the type of the genus Pseudogekko and suggested a possible relationship with Thecadactylus. The genus is very distinct and we are also somewhat in doubt as to its true affinities. However, comparison of this single specimen from the Solomon Islands with a paratype of Lupcrosaurus mac gre g ori and the description of the genera Thecadactylus and Pseudothecadactylus based on Brongersma's detailed consideration of their differences (1934, pp. 176-179, and 1936, p. 136) we are inclined to believe that it is probably more closely related to the genus Luperosaurus. The type of Pseudogekko compresicorpus was deposited in the collections of the Philippine Bureau of Science, No. 1781, (Taylor, 1944, p. 148). This and probably two embryos mentioned by Taylor 1922, p. 105) are to the best of our knowledge no longer extant due to the loss of the collections of the Philippine Bureau of Science during the second World Wair ; and these were, as far as we were able to find in the literature, the only known specimens of the genus. Our assigning of the present species to this genus is based on general similarities which are much greater than the aparent differences in the nature of the claw sheath as shown in Taylor's drawing (fig. 12 c, 1922, p. 104) and our own figure 1 which follows. Taylor's drawing suggests that the ventral portion of the claw sheath is formed by the two halves of the terminal lamella. In the present species as well as in Luperosaurus macgregori the ventral portion of the claw sheath is a pointed median plate which is frequently indistinguishable from the claw except in a lateral view. We venture that the same ventral sheath structure was present in the specimen of P. compresicorpus but passed unobserved. The fact that the terminal lamella in the present species is actually undivided is paralleled by the conditions in Lepido-