418 PROCEEDINGS OF THE ACADEMY OF [J^lty, 



represents growth. However, the writer found that the snails, large 

 and small, are mathematically similar. That is, the weights are pro- 

 portional to the cubes of the length. 



In determining the average weight all the snails from a given experi- 

 ment were placed on a microscopic slide of known weight and all excess 

 water wiped aw^ay with a clean handkerchief or with lens paper. They 

 were then let dry for three or four minutes and weighed. Dividing the 

 weight found for the snails by the number gives the average. 



Tables. — The results of experiments are placed in tabular form. 

 Each experiment consists of two parts ; the second member of the pair 

 is in every case the control or the condition most nearly normal. The 

 variable factor precedes it. 



A detailed description of the vertical columns of the tables will now 

 be considered. Cf. tables, pp. 421 et seq. 



Column 1. — The serial number of the experiment. 



Column 2. — The number of days the experiment was carried on. 



Column 3. — The number of snails placed in a jar at the beginning of 

 the experiment. 



Column 4. — The number of snails alive at the time the measure- 

 ments were taken. 



Column 5. — ^The condition that varied in each pair of experiments. 



Column 6. — Certain constant conditions. These constants are in 

 some cases interesting to know. In this column certain abbrevia- 

 tions are used. W P equals Water Plant, i.e., Myriophyllum, Cera- 

 tophyllum, Elodea or Spirogyra. N indicates no water plant, cc. 

 equals cubic centimeters present in each jar. 



Column 7. — Average size. If the number is expressed in ten thou- 

 sandths, grams are to be implied; if expressed in units and tenths, 

 millimeters. 



Column 8. — The differences between pairs are placed opposite the 

 largest number of the pair. 



Column 9. — The quantity in this column is the per cent, of the differ- 

 ence to the largest average of the pair. To be able to compare the per 

 cent, difference of the weights with those of lengths the following 

 formula was used, being based on the fact that the snails are similar. 



a and b are two members of a pair expressed in grams and a > b. 

 Since the shells are similar mathematically, then 



Analysis of Experiments. — To bring the mass of experiments into a 



