88 The Pi,ant Wori.d. 



involving a prolonged series of experiments, but the appearance 

 or non-appearance of both these and the summer annuals is so 

 obviously related to rainfall that any other conclusion as to their 

 relation to water supply is, with present evidence, impossible. 



The summer annuals appear after the summer rains, during 

 the period of the highest temperature of the year. Once the 

 ground has become wet they germinate and develop with great 

 rapidity, some of them growing to a large size and in such num- 

 bers as to form, in many places, a conspicuous feature of the vege- 

 tation. They bear the same relation to summer rains that the 

 winter annuals do to winter rains, and their behavior necessitates 

 the same conclusion as to the fundamental importance, in their 

 case as in that of the winter annuals, of the water relation. 



Not less certain, however, is the fact that the habits of both 

 are directly correlated with differences of temperature. It is 

 impossible, for example, to induce the germination of the seeds 

 of winter annuals in summer temperatures, unless they have 

 previously been subjected to a low temperature. Thus while 

 water supply, with annuals as with perennials, is a chief factor in 

 determining distribution in space, differences of temperature 

 determine with remarkable precision and certainty the distribu- 

 tion in time of the winter and summer annuals, and the face of 

 the landscape is twice a year changed to correspond with their 

 biological habits. It results that areas which are densely covered 

 with winter annuals in February and March afford ample room 

 and all the necessary conditions of development to the luxuriantly 

 growing summer annuals of August and September. 



(9) Parasitic and Symbiotic Plants. 

 Though not forming associations properly comparable 

 with the foregoing ones, the parasitic plants of the Laboratory 

 domain call for notice. Two species of mistletoe are widely 

 distributed in this region, one of which, Phoradendron calif or ni- 

 cum, is of fairly frequent occurrence on Tumamoc Hill and th 

 adjacent valleys, where it grows on Prosopis vekitina,Parkin' 

 sonia micropyhlla, Cercidium Torreyanum, Acacia constricta 

 and A. Greggii. Of these P. microphylla and the mesquite are 

 its most frequent hosts on the domain, where it has been observed 

 but once on A. constricta, and once on Cercidium Torreyanum. 

 The other species, Phoradendron flavescens, has not been found 



