MATERIAL AND METHODS. 15 



narrow (6 to 8 feet) channel. It lies on the south side of the river. 

 The bottom of this pond is covered with a dense mass of aquatic vegeta- 

 tion, principally consisting of Chara, but with a good deal of Cerato- 

 phyllum about the edges of the Chara banks. On the north side of this 

 pond, which is about a hundred yards wide, there is a thick growth of 

 willows and alders, which overhang the edge and give a considerable 

 amount of shade to the Ceratophyllum beneath. On the south side, on 

 the contrary, there is no shade whatever, the aquatic plants being 

 exposed to the full glare of the sun during the whole day. On the south 

 side the Ceratophyllum was closely associated with Chara, forming a 

 fringe about the Chara bed. On the north side, however, the Cera- 

 tophyllum was growing alone. In other respects the conditions were 

 the same on the two sides of the pond. Series V, including 7 plants, 

 was collected from the north side of this pond, and Series VI, including 

 6 plants, from the south side. The date of collection was May 21, 

 1905. The plants thus represent the early spring conditions. Nearly 

 all the plants were small. The water where the different plants were 

 taken varied between 6 inches and 2 feet in depth. The bottom was 

 physically very like what has been described above for the Carp Lake 

 habitat— soft black mud, with much plant debris. All the plants were 

 attached. The water is very quiet, as the pond is well protected from 

 the wind, and there is no appreciable current. The plants of these two 

 series were of special interest because of the fact that they included 

 representatives of each of the methods of passing the winter which 

 have been mentioned above (p. 10) . Plants 1, 2, 6, and 7 of Series V, 

 and plants 1, 2, 4, 5, and 6 of Series VI had wintered over unbroken 

 and had begun new growth in the spring without having lost their 

 individuality of the previous year. The other plants of these two series 

 had started in the early spring, either from seeds or from separated 

 winter buds. 



Putting all the series together, it will be seen that we have plants 

 from the beginning, the height, and the end of the growing season, 

 representing lake, pond, and river habitats. Our material must, then, 

 be regarded as fairly comprehensive. 



Finally, a word may be said regarding the calculations. I have 

 followed throughout the plan of calculating the standard deviation from 

 the unmodified second moment, wherever it was to be used simply as 

 an index of variation. In calculating the moments for fitting curves to 

 the observations I have tried Sheppard's corrections wherever there 

 was any approach to high contact at the ends of the range, but usually 

 with not very satisfactory results (cf . p. 23, infra) . In each case a 

 statement is made as to whether the moments were modified or not. 



