Dec, 1902,] Observations on the Dandelion, 335 



during another. No. One never grew one inch during any day- 

 time except once ; No. Two lengthened one inch in one daytime 

 and two inches in one other. No. One lengthened rapidly during 

 the last three days and nights just before it completed its growth. 

 No. Two did the same, except that it grew^ an eighth of an inch 

 after the seed had been scattered, probably owing to a shower. 

 No. One did not grow any for three days and nights just before 

 opening the head of pappus, but the weather was cold and rainy ; 

 No. Two lengthened very rapidl}- during the three days and 

 nights just before scattering the seed, probably because the 

 weather was warm and windy, with a slight shower the night 

 before the head opened. No. One did not grow any during the 

 three days and nights just after blooming, probably because its 

 strength had been exhausted in flowering ; No. Two did the 

 same. This was the longest period of rest from growth that 

 either plant took. The scape of No. One grew over an inch 

 during the days and nights the head was in bloom ; No. Two, 

 two inches. No. One grew one inch during the day and night 

 just before blooming ; No. Two, four and one-third inches the 

 night previous to blooming — its most rapid growth. 



In No. One, previous to blooming, the scape bent over to 

 lower and protect the head, but became erect the night before the 

 head opened. By this process the head was elevated over an 

 inch. No. Two did the same. Along the roadside this process 

 often makes a difference in elevation of the head of four inches, 

 especially if the neighboring vegetation is growing. After 

 blooming the scape of No. One again flexed to lower and protect 

 the head while the seed should ripen. By the scape's assuming a 

 compound curve the head stood erect. The scape became straight 

 and vertical two days before the .seeds were to be scattered, thus 

 raising the head an inch more than it otherwise would be. Along 

 the roadside this change in the direction of the scape often 

 elevates the head as much as six inches. The record for No. Two 

 was the same as for No. One. 



In No. One the corollas died and dried in about five da^-s after 

 blooming, and at the end of three more the growing jDappus 

 stalks pu.shed them off ; for No. Two no record was made. In 

 No. One the plant was five da5'S in reflexing the scales of the 

 outer involucre ; but after they were once down they never again 

 became erect. The same was true of No. Two. In No. One 

 the scales of the inner involucre curved outward near their middle 

 to permit the flowers to bloom, but became erect in the e\-ening 

 and at night to close the flower. They also remained erect while 

 the seed was ripening. When the head of pappus was ready to 

 come out this was accomplished by the receptacle's changing from 

 a concave disk to a convex one. This mechanical device not only 

 opened up the head of pappus, but reflexed the involucres at 



