1912.] 



NATURAL SCIENCES OF PHILADELPHIA. 



181 



ordinate. The index on Table I is the abscissa. On Table II the 

 ratio is the abscissa. Table I shows that the larger shells have a 

 narrower aperture, which agrees with H. B. Baker, but Table II 

 shows that the height of the spire is about constant at all ages. On 

 both tables no conclusions can be based on shells under eight milli- 

 meters, for two reasons : the numbers of shells are too small and the 

 probable error of each ratio is too large. It may be as much as ±.10 

 on the part of the index and ±.05 for the ratio. 



Remembering, then, that the larger shell may have a narrower 

 aperture, then we may look over fig. 3. In this the Cedar Lake 

 collection and the Wingohocking Creek collection both have narrow 

 apertures, but the former has an average shell of 9.44 mm., while 

 the latter has an average shell length of 17.37. The size of the shell 

 cannot influence, then, the characters in question in this case. The 

 other diagrams tell their own story and seem to show that each 

 restricted area has its own type of shell as far as these two characters 

 are concerned. The numbers are far too small on which to base 

 many conclusions. 



Fig. 4. 



There were .hatched during the past winter from an egg capsule 

 laid by a snail from the Sixty-seventh Street and Elmwood Avenue 

 collection thirty-two snails, of which eighteen lived to be measured. 

 These were kept together in 1,000 cc. of water in a crystallizing dish. 

 They were measured from time to time. From two of these meas- 

 urements, an early one and a late one, 37 days and 78 days, respect- 

 ively, fig. 4 was compiled. 



The probable error of the individual 37 days old measurements, 



