BOTANICAL SUBJECTS. 181 



In Lathyrus macrorhizus (Fig. 51) the tul3ers are from the 

 size of a pea to that of a small cherry, and would appear to be 

 enlargements of the voot-nodes. 



In most legiiminifera?, however, the tubercles are very minute, 

 varying from the size of a pin's head to that of a mustard seed, 

 and such as we find them in Vicia augustifolia and others 

 (Fig. 52). 



Curiously enough, the Lathyrus with large tubers does not 

 appear to possess the minute one,* so that the theory that the 

 latter are factories of nitrogen assimilation, through the low 



Fig. 52. Minute tubercles on roots of Vicia augustifolia (" Syme'sBrit. Bot."). 



organisms within them, would be very much weakened ; for we see 

 that the minute tubercles in the one are replaced by the large 

 tubers in the other (if this be so), so that one begins to think that 

 the minute tubercles may possibly be only atrophied representatives 

 of the more useful tubers. 



This question, however, naturally arises, why have the 

 leguminiferas tubers or tubercles at all, when so many other 

 plants have them not ? The foUomng answer suggests itself. 

 The flower of the leguminifer^e is one which has undergone 

 great modification. It is, therefore, conceivable that when it 

 began to modify, there was not present a sufficiently mochfied 

 insect, suited to fertilize its ovules. It had either to perish 

 or, so to speak, accept the help of such accidental storage of 

 energy in its roots. This accidental storage mav in time have 

 developed into what we now call tubers. In short, it had to live 

 and wait until something turned up by which its flowers could be 

 fertilized, and so produce further variations more suited to its 

 surrounchngs. "When seed began to be annually perfected by 

 insect agency the tubers atrophied by transfer of energv, but, never- 

 theless, they continued to be inherited as remnants^ much as the 



At all events they are not shown in Syme's figure ("Engl. Bot."). 



