OBSERVATIONS ON FLOWEKING PERIODS OE CERTAIN CACTI. 37 



observe any such behavior in plants in their natural habitat. The 

 flowers showing no greater inclination to open in the sunshine 

 than in the shade. 



The flowers remain open till about 9 :00 p. m., when the first 

 signs of closure are to be observed. A somewhat analogous be- 

 havior was recorded for ''Cactus grand if orus" bv Linnaeus in 

 his floral clock, who reported it as opening at 9 :00 or 10 :00 

 ]). m., and closing at midnight. Closure occupies some hours, 

 this being not a well-marked period. The earlier stages of clos- 

 ure are indicated by the upward position of the inner petals, and 

 the assumption of a rotate position by the remaining members of 

 the corolla which earlier Avere strongly recurved. Somewhat 

 further inward flexure of the petals is followed by a general 

 wilting of all its members, and these gradually wither and dry 

 up in position, the stamens being most resistant. The formation 

 of a scission layer is very rapid, so that in forty-eight hours the 

 whole floral apparatus readily falls away from the ovary. In- 

 deed, the point suggests itself if the rapid growth of the scission 

 layer may not be the immediate partial, if not the whole, cause of 

 the closing and withering away of these flowers, an idea which 

 is reinforced by the appearance of the petals, which, evanescent 

 as they are, are not at all resistant. On the other hand, there is 

 evidence that the scission layer in such flowers as those of Cereus 

 TTiurheri, C. giganteus, etc., is not formed rapidly, though this 

 may turn out not to be true upon examination. The behavior 

 of the scission layer as related to the period of flowering in the 

 cacti is one which promises a good deal of interest to the student. 



The rapidity of closure in the cholla, under the ordinary 

 conditions in the dry portion of the summer, is as just described. 

 There are, however, exceptional conditions under which the flow- 

 ers behave somewhat differently, that is, the closure is retarded. 

 Rain or a condition of high relative humidity following rain, 

 are favorable to a retention of the open condition, though I have 

 not found that these induce an earlier than normal opening, when 

 they are not accompanied by dense cloudiness. Thus on August 

 19, 1906, the flowers were still open at 9 :00 a. m., the parts 



