91 



stripe developed, while G. n. sp. from Funafuti has no grooves and G. nohilis n. sp. 

 (PI. II, fig. <Sa) has l\vo(7) almost oh.solete ones. Most species of Olpiiim L. K. show 

 no trace of transverse stripes; Ul. Birmanicum n. sp. and Olpinni (Jrtonedae, forms 

 which also in other respects stand apart, have a broad median stripe (PI. II, 

 lig. 7 a). It ought to be pointed out, that the presence of an indistinct stripe or 

 its complete absence is very difficult to determine; the result seems often to be a 

 matter of taste. 



Daday writes (11. p. 179) about the difference between his Garypinae Sim. and 

 Garypininae Dad. "Der letzteren Unterfamilie IGarypinae] ähnelt sie [Garypiiiinae] 

 aber auch dadurch, dass von den Abdominalsegmenten die drei ersten auf dem 

 Rücken ungetheilt sind, während bekanntlich bei den Cheliferinen jedes Abdominal- 

 segment getheilt ist." Several of the facts, set forth in the passage quoted, are 

 not quite correct; only very few Cheliferinae Sim. are found, in which the first 

 abdominal tergite is longitudinally divided; in most forms we find the first and 

 eleventh tergites at least undivided (cf. 46. Balzan's figure of Ch. segmenlidentatus 



Balz tav. XV, fig. 13 ^). His statements about the longitudinal line in the 



Garypidae is, as will appear from, what follows, not quite correct. All the abdo- 

 minal tergites except the last one, are longitudinally divided in Garypiniis Pala- 

 goniciis Ell. (70. p. 6), and so they are in G. nobilis n. sp., but rather indistinctly 

 and with the exception of the first and last ones; a good many have from the 

 fourth to the tenth tergites more or less distinctly divided by a longitudinal line 

 f. inst. Garypiniis dimidiatus L. K., G. n. sp. from Hawaii and Olpinm Birrminiciim 

 n. sp. Most species of Olpinm L. K. with the exception of the last mentioned and 

 a few besides have the abdominal tergites undivided. The longitudinal line in 

 species of Garypns L. K. is most often broad and indistinct and it is often difficult 

 to determine, if they are divided or not divided; Garypns latus Hans, has all the 

 tergites divided but the two last ones, while Garypns Floridensis Bks. has from the 

 third to the tenth divided. 



The size of the antennae seems to a certain degree to correspond to the 

 development of the cucullus; they are exceedingly small in forms, like the species 

 of Garypns L. K. in which the cucullus is well produced and the front margin 

 narrow, while they are larger in most forms of Olpinm L. K., in which the front 

 margin is fairly broad. The marginal teeth of the fingers are as a rule minute 

 and slightly developed. 



As far as the structures and the modifications of the flagellnm. lamina interior, 

 serrula exterior and galea, organs, which in spite of marked variations within genus 

 as well as species contribute to the characteristics of this family, are concerned, 

 I refer to the general part (pp. 12—22) as well as to several figures on plate II as 

 well as a few textfigures. 



The lahriim and maxillae have not so complicated a structure as in Chelij'cr 

 Geof. ; the former has no hindmost prolongation as Obisium 111. The median lyri- 



12* 



