436 THE AMERICAN NATURALIST [Vol. LV 



SO far noted in which the seeds of a single plant of apricus 

 are not entirely uniform is that of two tones in a striped 

 white variety. An attempt has been made to cultivate 

 the two sorts separately and to determine the ratio in 

 which they occur. Cultures have not yet proved success- 

 ful. In some collections of seed the ratio approximates 

 a 3:1, but in others it is nearer 8:1, but whether any 

 significance is to be attached to these results has not yet 

 been definitely determined. 



It is a curious fact that, although lupines yield an 

 enormous number of seeds and the plants often literally 

 cover acres of space they are nevertheless very capricious 

 in their occurrence. Some illustrations of this may be 

 taken at random from our notes. The field near the 

 university where the apricus mutants and the pink piper- 

 sniithii were collected in 1914 has been under observation 

 every year since and has not at any time had many plants 

 or produced a second display of these mutants. A num- 

 ber of pink forms of both species were staked for seed 

 that first year and the stakes left in the ground to mark 

 the site, in the expectation that the same forms would 

 reappear the following year. In no single case were pink 

 flowers found at any of these stations, although they were 

 found the next year at other locations in the same field. 

 The field near Woodside studied in 1917 lias Ixni visited 

 each year since. Not only have no mutant foi nis hcon 

 found there, but there have been exceedingly few normal 

 ones. The field from which the notes were made for the 

 table on a preceding page was very thoroughly searched 

 again this spring. There were a few dark blue plants, but 

 not a single one of the types which were more or less 

 abundant there last year. These vagaries of distribu- 

 tion doubtless depend in some manner not yet clear on 

 the difficulty of germinating the seeds. 



Pnllniafinu aurl Seed Collection.— In 1914 when these 

 ol)>('i \'at!(>ns wcic begun it was assumed that the lupines 

 well' i-robal)!;. I'fcfiuently cross-pollinated, inasmuch as 

 they appeared to be freely visited by bees. It was a 



