290 



ridge there will be a general thinning down of the adjacent 

 tissues. Where this thin stratum occurs, the plant will 

 probably encounter much less than the average amount 

 of resistance, and may feel disposed to make use of this 

 easier outlet for the development of sori. The fact that the 

 sori in the marginate section are often confined to one long 

 continuous narrow line stretching the whole length of the 

 frond on each side of the midrib, instead of being arranged 

 in the usual double set of transverse lines, lends a certain 

 amount of plausibility to this hypothesis. 



There appears to be much difference in the times of 

 ripening and shedding of spores. In many instances we 

 may be able to gather ripe spores almost anytime from 

 June until December. There is, however, at least one 

 notable exception in Osmunda japonica and its varieties. 

 The time of ripening and consequent shedding of spores of 

 this species comes like a thief in the night. As the critical 

 time approaches I have tapped and examined the fertile 

 fronds several times a day without being able to extract any 

 ripe spores. At the next examination, in perhaps half-an- 

 hour, I have found every shell practically empty and the 

 spores all shed. The only safe plan is to place a large 

 sheet of white paper under the plant to catch the spores, 

 and examine it periodically. 



The muricate section of Scolopendrium has, next to the 

 marginate, given us the greatest number of examples of 

 suprasoriferation. Examined through a microscope, the 

 upper surface ot a muricate Scol appears to be pitted with 

 shell craters like a miniature representation of the battle 

 field in France. Where these craters are deepest, the hollows 

 will be softer and more tender than the upper surface in 

 normal types. The sori will be pushed out along the lines 

 of least resistance, and in the roughened Scols the least 



