SURFACE COPEPODA OF THE GTJLF OF MANNAR. 199 



present in Calanus minor. As I have previously mentioned 

 {vide ante, p. 192), I consider that these forms are in all 

 probability examples of Calanus minor. 



As A. Scott {loc. cit.) has pomted out, the form known as 

 U. caroli $ is easily recognizable from U. darwini ^ by its 

 constantly smaller size and by the structure of the fifth |)air 

 of legs. 



In a tow-netting taken at " Investigator " Station No. 470 

 (Sawi Bay, Kar Nicobar, December 7, 1912), large numbers 

 of both forms were present, associated with numerous examples 

 of Undinula darwini $ ; a comparative study of the two 

 forms of male has given the following interesting results. 



The proportional, lengths of the abdominal and fm'cal 

 segments in the two forms are — 



Undinula darwini : 45: 41: 32: 27: 25: 23. 

 Undinula caroli : 45 : 41 : 31 : 27 : 24 : 22. 

 The proportional lengths of the antennal joints are as 

 follows : — * 



Segments: 1-2. 3-5. 6. 7. 8-9. 10. 11. 12. 13. 14. 15. 



U. darwini .. 121 82 26 27 40 26 31 38 46 46 49 

 U. caroli ... 118 79 24 26 39 28 30 37 45 45 50 



Segments : 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 



U. darwim .. 49 51 53 57 46 46 51 51 46 18 

 U. caroli .. 50 51 53 56 47 47 53 53 50 19 



The structure of the first four pairs of swimming legs is 

 exactly similar in the two forms, not only as regards the 

 spinulation, but also in the proportional lengths of the various 

 segments ; thus the proportional lengths of the segments and 

 end spine in the third leg of the two forms are as follows : — 



* In all cases the total length of the antenna has been taken = 1,000. 

 In this way it is found that comparison between specimens of luxequal 

 size is greatly facilitated. 



