MATERIAL AND METHODS. 13 



plant studied, it will be seen that they give more complete data for 

 the study of positional differentiation than probably has hitherto been 

 available for any plant. 



Our material came from three localities and is representative of four 

 distinct habitats. It is comprised in six series, an account of which 

 follows. 



Series I, II, and III. —The plants composing these series were collected 

 in the summer of 1902, from Carp Lake, near Grand Traverse Bay, in 

 the northern part of Michigan. Carp Lake is a long and narrow body 

 of water extending about 18 miles in a generally north and south direc- 

 tion, with an outlet into Lake Michigan at its northern end. At its 

 narrowest part it is only about a half mile in width. At this point there 

 is a neck of marshy land extending out into the lake from the east side 

 and reaching more than halfway across it. On the north side of this 

 strip of land (known locally as Fountain Point), the plants of Series I, 

 II, and III were collected. They all came from a small open space sur- 

 rounded by an abundant growth of cat-tails and reeds (Juncus sp.). 

 The situation in which Ceratophyllum was growing was well protected 

 from the wind, and consequently the water about the plants was rarely 

 much disturbed. The plants were directly exposed to the sunlight, there 

 being no overhanging trees or bushes. The bottom was similar to that 

 usually found under Ceratophyllum beds. The proximal ends of the 

 plants were buried in soft, loose, black mud, containing much decayed 

 plant debris, which came mostly from the Ceratophyllum itself. The 

 growth of the plant at this place was very abundant. The individual 

 plants were so matted together in the main mass composing the bed 

 that it was only with difficulty that a single one could be disentangled 

 unbroken. 



The dates of collection for the three series were as follows: 



Series I: Collected July 22, 1902. At this time the water at the 

 point of collection was 2 feet deep. 



Series II: Collected August 18, 1902. Water at point of collection 

 approximately 15 inches deep. The level of the whole lake had lowered 

 since the first collection. The water surrounding the plants was very 

 muddy. 



Series III: Collected August 25, 1902. Conditions were the same 

 as when the Series II collection was made. 



Series I included 5 plants. Series II, 2 plants, and Series III, 1 plant. 

 The reason why Series II and III did not include larger numbers was 

 the great difficulty of collecting unbroken specimens after the water 

 had become so low. The matting together of the plants in the bed was 

 so close as a result of this lowering that to disentangle a large plant 



