214 DISCOVERY REPORTS 



would enable growth and moulting to be free of the restricting egg-carrying condition 

 for long intervals. 



The spawning time of the second species, M. suhnigosa, is less easy to determine. The 

 graph (Fig. i) showing the percentage of berried females in this species indicates a 

 possible shedding of larvae in February ; for after climbing to a pronounced peak in that 

 month it shows a sudden drop to zero in March which persists in April and May. The 

 paucity of Munida larvae in the plankton during February (three present in two hauls 

 out of six) and their total absence in March and April makes spawning at this time of the 

 year unlikely. No date, therefore, can be given for the spawning of M. subrugosa, but 

 the likeliest time for this event would seem to be in early summer, in September and 

 October, as in M. gregaria, a time when large numbers of Munida larvae are present in 

 the plankton. Eggs obtained from ovigerous females taken on September 20 and 

 October 13 contained well-developed embryos, but those of females captured on 

 September 5 showed no indication of an embryo. It may be then that the spawning 

 time of M. subrugosa is more variable than that of M. gregaria, and that larvae are shed 

 into the plankton all through the summer. 



In the series of larval Munida which has been examined it has not been possible to 

 recognize two forms to correspond with the two species that are to be found in the adult 

 condition in the area, and it is possible that the differences between the larvae of these 

 two species are so small as to be imperceptible. Certainly the differences between the 

 very earliest post-larvae, before the modification of the external maxillipede in M. 

 gregaria, are such as to make distinction at this stage difiicult or even impossible, and 

 considering the close resemblance and relationship of the two species, a very marked 

 similarity or even morphological identity of the two larval forms would not be sur- 

 prising. 



The Falkland Islands Munida larvae resemble the other described forms to a marked 

 degree, but they can be distinguished by a different spinulation of the posterior dorsal 

 margins of the abdominal segments. Five larval stages are here described, differentiated 

 primarily by the spines of the telson. 



STAGE I (Fig. 2«) 



The two spines on the posterior dorsal margins of the second to fifth abdominal 

 segments in the first stage of M. banffica are replaced in the Falkland larvae by four 

 prominent spines, in the spaces between which are two smaller spines. Laterally to 

 the outer larger spines are one or two smaller spines, making twelve to fourteen in all. 

 The fourth and fifth abdominal segments each carry two strong lateral spines exactly 

 similar to those present in M. banffica. The first abdominal segment carries a row of 

 very fine spinules on its posterior dorsal margin, a feature common to the more 

 posterior segments of the larvae of Galathea. 



This is the stage figured by Gurney (1924) from a specimen taken north of Three 

 Kings Islands, New Zealand. His figure shows features very much like the Falkland 



