GRADATION 203 



continuity by which two successive periods (waves) of gradation 

 are disjointed. 



The existence of secondary humps (or breaches of continuity) 

 in a gradation curve of a given specimen depends probably very 

 often on a temporary change in the external conditions (tem- 

 perature ? supply of water ? etc.) in the course of the period of 

 growth. In certain species, however, the existence of more 

 than one period is observed in all the specimens ^ and is there- 

 fore characteristic of the species. This is the case with the 

 gradation of the primordia length and breadth of the leaves of the 

 fertile stem in Cinclidium stygium Sw. (a moss allied to Mnium). 



Fig. 27.— The curves represent the figures on p. 202 . a, length of the inter- 

 nodes ; h, length of the sheaths ; c, length of the blades ; d, breadth of the 

 blades ; I, V, . . . successive intemodes. Scale in millimetres 



Further investigation (by measurement !) of this subject may 

 bring, I think, an explanation of a number of facts in the 

 structure of living beings which have been hitherto overlooked 

 or looked upon as being mere curiosities. 



In the gradation curve of the length of the sheath (Rolens) one 

 observes a distinct maximum (hump) in the upper leaf (inter- 

 node XVI.) and a secondary hump in intemode X. The latter 

 did not exist or was hardly perceptible in other stems of the 

 same species. The gradation curves of the primordia len^h 

 and breadth of the blade have a rather regular ascending- 

 descending form, their summits coinciding respectively with 

 the ordinates XII. and XIV. 



'The gradation curve of the property under consideration being simple 

 (one-humped) in allied species. 



