158 Memoirs of the Indian Museum. [Voi,. IV, 



and small specimens and, in the latter, they are sometimes quite separate from the 

 median keels and so form incomplete additional carinae between the medians and 

 laterals. The distal end of the median keel may be rounded, bluntly pointed or sharply 

 spinous either with or without the addition of anchor-flukes. The flukes themselves, 

 in cases where they are separate from the median ridge, may also terminate in spines. 



The three main keels may take the form of thin narrow crests or may be greatly 

 swollen, but in practically every known instance there is a definite interspace between 

 them, and between them and the keels on the marginal teeth. The submedian and in- 

 termediate marginal teeth are well-developed and each is supported by a prominent 

 longitudinal ridge. There are as a rule numerous spinules on the inner edges of the sub- 

 medians and a small rounded lobe or denticle in conjunction with one or two spinules 

 between the submedians and intermediates. At the tip of each submedian tooth there 

 is a small movable spinule. The lateral margin is sharply carinate anteriorly and 

 terminates in a sharp notch representing the lateral tooth. This tooth is never well- 

 developed, and in some cases is perhaps wholly absent. 1 



The basal segment of the uropods is provided with a short dorsal spine projecting 

 over the first segment of the exopod. Inferiorly it is produced to a long broad process, 

 divided by a distal emargination of less than half its entire length into two strong teeth. 

 The outer of these reaches almost or quite to the apex of the exopod and is usually a 

 trifle longer than the inner. The internal margin of the outer spine is unarmed and 

 does not bear a small tooth as in large specimens of G. glabrous and graphurus. The 

 basal segment of the outer uropod projects a little beyond the articulation of the ultimate 

 segment, and bears, as a rule, ten or eleven movable spines — the outermost sometimes 

 slightly recurved at the apex — on its external edge. The inner uropod is oval and 

 elongate and, as is customary, bears a single row of setae on its margins. 



In the table on p. 159 I have attempted, with the help of the key given by Borra- 

 daile (1907, p. 211), to arrange the 224 specimens that have been examined under differ- 

 ent varieties, giving the sex and approximate length in mm. of each specimen. The 

 form K ( = ' confinis ') is not represented in the collection. 



Owing to the fact that the series is perfectly continuous in its variation it has 

 proved well nigh impossible to catalogue the various individuals in a satisfactory 

 manner, but conclusions of some interest are none the less indicated. 



The table will, I believe, be found to substantiate my statement (p. 150) that 

 young examples exhibit a much greater range of variation than those of large size. 

 This is well shown in the case of the specimens referred toH( = ' afftnis '), and is also 

 noticeable in those catalogued as HA, AD, D, HD and HK, while some corroboration 

 is afforded by a study of the data given by I v anchester. Of the three explanations 

 which I have suggested the most probable in my view is that the young forms alter 

 during subsequent moults and eventually reach or nearly reach the standard adult 

 pattern ; it is possible, however, that this is not the case with D ( = ' smithi '). 



In the next place it will be seen at a glance that A ( = 'incipiens ') is far more 



1 It is present in every example of G. chiraçra that I have seen, but is absent in all specimens of the 

 var. platysoma. 



