1908.] NATURAL SCIENCES OF PHILADELPHIA. 441 



in a larger volume. This coidcl be explained by assuming that there is 

 but a definite amount of toxin present to act on the seed. With the 

 snail, however, the case is reversed. The toxin, which we have shown 

 in the preceding sections to be present, is ever being increased in 

 quantity by the secretions of the animal. In the case of the seedling the 

 solution becomes weaker and weaker. 



Pearl and Dunbar ('05) found that Paramecium in small vessels were 

 dwarfed. This is due most likely to the accumulation of excreted 

 matter. In fact almost every case of this kind among aquatic animals 

 can be so explained. 



8. Alternation of Conditions. — In connection with some of the experi- 

 ments on heat and cold, a jar was moved from the warm to the cold, and 

 vice versa, at two weekly intervals for a period of two months, with 

 the very striking result that the alternated snails were larger at the 

 end of that time than those kept in the warm all of the time. This 

 result was accomplished notwithstanding the fact that, when in the 

 cold, the water in the alternate jars was sometimes frozen. This 

 experiment led to a series of experiments in the same line, and although 

 many were as striking as the first, yet the larger snails were those, as a 

 general rule, that had been in the warm room all the time. This con- 

 trol in the warm room was every two weeks transferred to a jar from the 

 cold conditions, while at the same time the jar in which they had been 

 living was placed in the cold and snails that had lived in the cold all the 

 time added. This process of changing the snails was performed every 

 two weeks or every week. The interval of alternation is given in the 

 tables. See Tables VIII-XX. Not only were alternate conditions 

 of heat and cold considered, but also alternating conditions of starving 

 and feeding and light and dark. The latter experiments are not of 

 particular interest, as the alternated snails are purely intermediate 

 in size between those under favorable and those under unfavorable 

 conditions. The starving and feeding experiments, however, closely 

 approximated those of heat and cold. Some were larger and some 

 were smaller than the control. These results must mean that the 

 change from an unfavorable to a favorable condition causes the snail to 

 grow faster than if it were continually in the favorable condition. 



9, Experiments on Tadpoles. — As Yung ('85) performed some experi- 

 ments on the effect of external conditions on tadpoles, arriving at the 

 same conclusion as did Willem ('96), i.e., that dwarfing was caused by 

 lack of aeration, the writer, using the methods described in the preced- 

 ing pages, repeated these experiments with tadpoles of Rana in the 

 spring of 1907. 



