1898.] NATURAL SCIENCES OF PHILADELPHIA. Ill 



semi-dark room, in a narrow-necked bottle filled with water from 

 the native spring. Evaporation is slow and distilled water only is 

 to be used in compensation. Any addition of town water is apt to 

 be followed by a sudden and enormous increase of bacteria and a 

 rapid decay of the diatoms. Bacteria do, in any case, gradually 

 invade the culture and ultimately destroy it; but, with the above 

 very simple precautions, it is quite possible to keep the gathering in 

 good condition for a month — long enough to observe a variety of 

 changes. 



Free access of oxygen promotes the production of the motile frus- 

 tules. Two approximately equal quantities of healthy filaments were 

 selected, one of which was placed in a little vial with a narrow neck, 

 the other in a flat glass dish ; and, other conditions, as temperature, 

 light and amount of water, being identical, at the end of a week the 

 filaments in the dish had entirely disappeared, and the glass was 

 found to be coated with single cells and small motile segments of the 

 original filaments, while the bulk of those in the vial remained un- 

 separated. 



A second experiment was as follows : — Two quantities of the un- 

 separated filaments, as nearly equal as may be, were put into pre- 

 cisely similar bottles. The first was at once placed in a dark corner 

 of a poorly lighted room, the second beside it, after having been ex- 

 posed to bright daylight until a mass of oxygen bubbles had formed. 

 In the course of twenty-four hours, during nine of which all light 

 was absent, the exposed bottle contained quantities of active motile 

 cells, while the other afforded only a very few. This experiment, 

 twice repeated with the same result, taken in connection with that 

 which precedes, would indicate that we have here to do with a pro- 

 cess the reverse of the endothermic chlorophyl reaction: — 



C0 2 + H 2 = CH 2 + 2 



and that the chemical side of the phenomenon is that which accom- 

 panies protoplasmic motile activity in general. 



This conclusion is reenforced by the observed fact that filaments 

 of Eunotia in process of separation, and the resultant motile cells, 

 give a strong carbon dioxide color-reaction in thoroughly aerated 

 water tinted with hematoxylin ; it is in accord with the important 

 observation of O. Miiller 1 as to the stimulating action of oxygen upon 

 Pinnularia ; and, finally, it is diametrically opposed to that view of 



1 Berichte der Deutsch. Botan. Gesellsch., Bd. XI, p. 571. 



