512 MacDougal : Svi^ir.iosis a\d Sapkophytism 



As a consequence of the above limitations all species furnished 

 with niycorhizas, tubercles, or which enter into direct mechanical 

 or nutritive associations must be classed as symbionts, or if it is de- 

 sn-able to maintain connection with existing literature dealing with 

 these forms it would be permissible to refer to them as symbiotic 

 saprophytes, although such designation must be regarded as ten- 

 tative, and justified by expediency only (i i). 



It is a matter of common knowledge that seedlings are holo- 

 saproph)'tic in the stage in which they are wholly dependent upon 

 the reserve material of the seed or fruit, and during the whole 

 period previous to the formation of chlorophyll in general. This 

 period is practically obliterated in those species in which chloro- 

 {)hyll is formed in the seed. On the other hand, the development 

 of this tendency has been twofold. The increase of the capacity 

 for the absorption of organic products has played an important 

 part in the reduction of certain seeds to their present minute form, 

 and again the retention, or extension, of this capacity throughout 

 a greater or less portion of the life of the sporoph}-te has resulted 

 m varying stages of true saprophytism, complete in one species of 

 the higher plants only. The duration of the holosapropiiytic stage 

 shows very wide variations in different species. In certain arums 

 it extends o\'er two years under natural conditions, and may be 

 extended by cultural methods so that the seedling miiy not form 

 chlorophyll until the third or perhaps even the fourth year of ex- 

 istence, according to experiments now in progress. Only those 

 species which show a marked capacity for the absorption and use 



cranic 



phyte should be classed as hemisaproph)-tes. Tlie henn'sapro- 

 phytes would consist chiefly of the carnivorous plants. The 

 greater majority of the species now included in this category are 

 in fact more or less symbiotic by means of mycorhizas, tubercles 

 or other nutritive associations. 



This paper describes an extension of the investigations detailed 

 in a previous paper (i i), and an effort has been made to determine 

 whether such associations constitute a single physiological type, or 

 wliether several types of nutritive adjustment are presented. 



