BIOTYPES AND HYBRIDS. 39 



group, as described, would include B. dp. heteris and B. bp. rhomboidea, while 

 the latter would contain B. bp. tennis and B. bp. simplex, and the ratio of the 

 two groups should be 3 : 1. The observed ratio was 187 : 66 or 2.83 : 1, a 

 fair agreement with expectation. The quantitative relations of the sub- 

 groups can not be derived from the notes, but among i !46 of the dominant 

 group discarded at one time, 25 were considered B. bp. rhomboidea, and the 

 most of the remainder were intermediate between B. bp. rhomboidea and^. 

 bp. heteris. According to theoretical considerations there should have been 

 36 B. bp. rhomboidea in that number. I have no doubt that this discrepancy 

 was due to the fact that too narrow a view was taken of the fluctuations 

 normal to the several recognized forms, so that some which were consid- 

 ered intermediate between B. bp. heteris and B. bp. rhomboidea were in 

 reality extracted B. bp. rhomboidea. This is even more certainly true with 

 respect to B. bp. heteris, for in the same group of 146 dominants, only 2 or 

 3 were recorded as B. bp. heteris. All those having" some incisions on 

 the proximal margin of the primary lobe were considered intermediate, 

 but pure-bred B. bp. heterish&s since been observed to possess these incisions 

 frequently as a fluctuating- character. No B. bp. simplex was recognized, 

 as this elementary species would have been thrown with B. bp. tennis with- 

 out question on the basis of depth of sinus and form of the terminal lobe, 

 which characters alone were used in the classification. 



054.28 : This plant was a good specimen of B. bp. rhomboidea, having 

 the incisions on both proximal and distal margins of the primary lobes, 

 and all the lobes on the earlier climax leaves rounded or rhomboidal. 

 The later rosette -leaves had some of the secondary lobes acutish, but not 

 elongated. The seeds were sown on April 18, 1906, and 325 plants were 

 potted. Of 302 which were studied later, 202 were classified as B. bp. 

 rhomboidea and 100 as B. bp. simplex. This is a rather large departure 

 from the expected ratio of 3 : 1, but here ag-ain the excess of B. bp. simplex 

 may be due to the fact that the characteristic marks of B. bp. rhomboidea 

 tend to disappear under unfavorable conditions of culture, so that some 

 that were classed as B . bp. simplex may have been modified specimens of 

 B. bp. rhomboidea. 



054.29 : This plant was a well-marked specimen of B. bp. heteris , 

 having slight denticulations on the primary lobes. It differed from the 

 usual habit of Bursa bursa-pastoris in having the leafy portion of the stem 

 absent. The seeds were sown April 18, 1906, and produced over 400 

 young- plants. These were readily divisible into two groups B. bp. heteris 

 and B. bp. tennis in the ratio 319 : 95 or 3.36 : 1. A few of the speci- 

 mens of this family which were classified with the dominant form differed 

 from B. bp. heteris in the lack of the rounded secondary. They were like 

 it, however, in that the sinuses extended completely to the midrib, thus 



