466 Mr. E. T. Pocock on 



line a little greater or less than or about equal to half the 

 width between the spines, the difference depending upon the 

 greater or less development of the lateral wings which support 

 the lateral spines ; the anterior border on level with the eight 

 anterior sigilla widely convex, truncate mesially, posterior 

 border immediately behind the posterior line of nine sigilla 

 straight or slightly convex, with tlie convexity looking back- 

 wards ; lateral wings nearly parallel-sided, the spines some- 

 times nearly following the direction of their respective borders, 

 but always diverging, sometimes a little, sometimes consider- 

 ably, the anterior looking forwards and outwards, the posterior 

 backwards and outwards, the border between them less than 

 the length of the posterior spine, and almost always greater 

 than that of the anterior, straight or directed inwards and 

 backwards ; posterior lateral spine twice or more than twice 

 the length of the anterior lateral ; posterior spines longer or 

 about as long as anterior lateral, separated at the base by a 

 space equalling from two to three times their own length and 

 equal to or a little less than the length of the posterior lateral 

 spine. 



Measui-ements in millimetres (of two specimens, A and B). — 

 Total length of abdomen along middle line in A 7, in B 5 ; 

 between lateral spines in A 12*5, in B 12 ; from tip to tip of 

 lateral spines in A 16 5, in B 18. 



Log. New Georgia {H.M.S. ' Penguin ' : type). Specimens 

 also procured by Mr. Woodford, probably in Shortland Island. 



The structural variations in this species do not appear to 

 be attributable to age nor to be correlated with geographical 

 distribution, each one of the series of seven collected by 

 Mr. Woodford being different from the rest; the same is 

 true of the three collected in New Georgia. 



There is little doubt that this species is identical with 

 G. moUusca of Keyserling (' Die Arachniden Australiens,' 

 1886, p. 88, pi. vii. fig. 1) ; but it is to my mind quite distinct 

 from the O. mollusca described and figured on p. 7, pi. i. 

 fig. 4, of the same work sixteen years earlier. Keyserling, I 

 think, confounded two species together when he described 

 what is here named G. signifer as the adult of G. moUusca. 

 At all events, none of the specimens of signifer, whether 

 young or old, that I have seen agree with the figure and 

 description of the original G. mollusca ; and in Keyserling's 

 collection of spiders, one of the specimens labelled by hira 

 G. mollusca is a representative of a form very like G. signifer ^ 

 while the others are referable to a species closely allied to 

 G. Westringiif Keyserling. 



