THE OSTEACODA. 



273 



CONCHOECIA ROTUNDATA, MuUer. 



This species is so small that my comparatively rough measurements, to the tenth of a 

 millimetre only, fail to bring out clearly the successive stages. Between the lengths 

 of 0'4 and I'O mm. there are (apparently) four stages, or three moults within a range 

 of 06 mm., and it is obvious that a finer method is required, which was not foreseen at 

 the beginning of the investigation. It is probable, however, that even if measurement 

 to the second place of decimals had been attempted, an overlap of the highest and 

 lowest extremes of successive curves vrould have occurred (compare the case of curta, 

 pp. 260-1). 



Fig. H. 



min 



o" 



100 80 



60 40 



20 



20 



40 



[00 



80 



60 



40 



20 



20 



40 



\Z 



I I 



I 



09 



0-8 



07 



06 



05 



04 



03 



IT ^ 



m 



w 



T t J- 



n n 



III 



in 



IV 



IV 



E 



IE 



IV 



N 



The left-hand figure summarises the actual observations ; in the right-hand figure an attcmi)t has been made to 

 resolve those ordinates which arc due to the overlapping of curves (p. 274). 



It is fairly obvious from the diagram that there are four stages on the female side of 

 the scale with maxima at ( + ) O'-l, 0'6, 0'8, 1-1. Similarly, anatomical investigation 

 showed that the amorphous curve on the male side was really soluble about four maxima 

 at ( + ) 04, 0'6, O'S, I'O; the divisions (moults) coming somewhere near 0"45, 0'7, 0"85. 

 The high ordinate at O? on the male side is therefore apparently due to the sum of 

 larger specimens of Stage III., and smaller specimens of Stage II. (low^er sexual stage) ; 

 unfortunately no record was kept of the numbers of these two stages at this length, 

 as the outcome of the investigation was not foreseen at the time. Consequently, only 



