Mr. H. Thompson. On Con-elation of [Mar. 1. 



common shrimp from Plymouth to be only O'll. Perhaps this 

 difference may in part be accounted for by the fact that the prawn is 

 essentially a swimmer, while the shrimp, in confinement at any rate, 

 spends most of its time standing or crawling on the surface of the 

 sea-bottom. The prawn is certainly a more powerful and rapid 

 swimmer. It swims backwards, propelling itself by means of the 

 hinder part of the abdomen, the terminal portion of which is fan- 

 shaped, the fan consisting of the telson and the exopodites of the 

 last pair of appendages. This hinder half of the abdomen starts 

 from the extended position, and is brought, with a rapid stroke, up 

 under the fore part of the abdomen and thorax, pushing the water 

 before it, and so propelling the animal backwards. The 3rd seg- 

 ment acts as the hinge of the propeller, and it may be assumed that 

 the exoskeleton is here exposed to the greatest strain, for here we 

 find the highest degree of correlation, as is shown by the following 

 figures : 



Index of correlation between terga of abdominal segments 1 and 2 = 0*58. 



2 3 - 070. 



3 4 = 071. 



4 5 = 0-62. 



5 6 = 0-57. 



6 and 

 telaon = 0-51. 



Professor Weldon determined three other values in the shrimp, 

 viz., the relations between carapace length and post-spinous portion of 

 carapace, between carapace length and tergum of abdomen vi., and 

 between carapace length and telsou. Their values were (in Ply- 

 month specimens) 0*81, 0'0'J, and 0'18 respectively. It is not 

 possible to institute a comparison with exactly the same organs in 

 the prawn, as, owing to the great variation in the length of the 

 rostrum, no appreciable degree of correlation exists between the 

 whole carapace including rostrum (measurement No. Ill) and the 

 other organs. But if we take measurement No. II (from orbit to 

 hinder edge of carapace) as an approximate equivalent to the cara- 

 pace measurements of the shrimp, we find in the prawn the corre- 

 sponding values are 0'59, 0'59, and 0'40. The difference in the first 

 measurement, viz., shrimp, 0'81 ; prawn, 0'59, may be accounted for 

 by the fact that the dorsal spines probably do not correspond in 

 nature or function in the two animals. The shrimp has but one 

 median dorsal spine, situated far forward ; the prawn_ has a row of 

 them. 



But why the prawn shonld exhibit a so much higher degree of 

 correlation between the carapace and the two terminal segments 

 of its body than the shrimp is not, I think, readily to be ex- 



