364 



K. C. PUNNETT. 



such a case the variability of the species as measured by the ordinary methods \vill be low 

 even though the range of variation is considerable. With the view to throwing some light 

 on the character of such variation among the Branchiostomidae certain measurements were 

 made on 28 specimens of Asymmetron lucayanum and 45 specimens of Heteropleuron maldiv- 

 ense. The results are given in Tables I. and II. It is to be regretted that a larger amount 

 of material was not examined by Mr C. Forster Cooper, to whom I am indebted for the 

 measurements recorded for A. lucayanum and H. maldivense. 



A glance at the two tables is sufficient to shew that the amount of meristic variation 

 in these two species of Cephalochorda is considerable. And not only the range of variations 

 but also their amount as measured by the standard deviation would appear to be large. 

 Observations on the meristic variability of the various groups of Vertebrata are unfortunately 

 at present very scarce, but such as are available for comparison with the Btanchiostomidae 

 are given in the appended table. 



TABLE III.' 



Asymmetron lucayanum... 

 Heteropleuron maldivense 



Spinax niger (5 adults) 



» >i (o I) ) 



„ „ ((J & ? embryos). 

 ScyUium canicula 



Pleuronectes platessa 



Clypea harengus 



Necturus maculatus. 



Locality 



Number of 

 specimens 



28 

 45 



144 

 84 



168 

 63 



341 



200 



170 



120 



254 



197 



47 



69 



61 



52 



53 



54 



100 



' The following customary abbreviations have been made 

 use of : — 



(r= Standard deviation 



_ / 2(»^/) 



P.E.ff = Probable error of standard deviation= ±^6745 -7=^ , 



>/2n 



C.V. = Coefficient of variability = 



ff xlOO 



where il/=the Mean of the number of segments and ?! = total 

 number of specimens. 



For a simple explanation of these formulae vide Pearson, 

 The Gravimar of Science, 1900, p. 384 et seq. 



