122 THE AMERICAN iMOXTHTT fAj.ril 



The uuinber of diatoms in tht^ origiual sainjilf is shown 

 by the broken line. The point at which this broken line 

 cuts the curves may l»e called the limit of p:ro\vth. In 

 Lake Cochituate this point was at a depth of about 

 twelve feet, in Chestnut Hill Reservoir, six feet. 



Diatoms are said to be positively heliot r(»i>i(', that is, 

 they tend to move towards the li^ht. In some species 

 this power is quite strong; in others it is less noticeable. 

 For the purpose of determinin<; the ludiotropism of the 

 diatoms commonly found in water supplies, samjdes of 

 water rich in diatoms were placed in brass tubes three 

 inches in diameter and thirty-two inches long, having glass 

 ends. One end was covered with a black cap, and the 

 other end e.Kposed to the light. After varying lengths 

 of exposure, })ortions of the water were drawn from each 

 end of the tubes and examined microscopically. As an 

 example of the results obtained the following may be 

 quoted. Cochituate water containing 1)22 diatoms per cc. 

 was exposed in a tube foi- twelve hours. At the end of 

 that time the water at the light eud of the tube contaiued 

 1,438 and that at tlie dark end only 320. Some of the 

 tubes were inclined, to see it' the diatoms would move 

 upwards towards the light; some of them were placed 

 vertically; in otheis the diatoms were given time to settle 

 before the exposure was made. The exjieriments showed 

 that most of the common genera tended to move towards 

 the light while settling, but that having once reached 

 the bottom of the tube they remained where they fell. 

 They apparently did not {)ossess the power of moving 

 upwards towards the light — certainly not thiough any 

 great depth of water. But while they could not rise of" 

 their own accord, slight currents of convection caused b\- 

 varying the temperature of the water sufficed to keeji 

 them near the surface. 



'i'he bearing which these facts ha\(' ujion the seasonal 

 (listribut ion of iliatoins is ol)vions. ninl we are now bette)- 



