EMBRYOLOGY IN RELATION TO TAXONOMY 371 



both: (1) divisions of the microspore mother cells of the simul- 

 taneous type; (2) generative cell cut off exactly opposite to the 

 furrow in the pollen grain; (3) ovule crassinucellate with two integu- 

 ments and an aril; (4) megaspore mother cell separated from the 

 nucellar epidermis by parietal cells; (5) embryo sac of Polygonum 

 type; and (6) endosperm of the Helobial type (demonstrated in 

 Gasteria and Kniphofia and inferred in Aloe from the position of the 

 primary endosperm nucleus). The more important differences 

 between the Asphodelinae-Aloinae and the Anthericinae are pre- 

 sented in Fig. 202. 



Liliaceae-Lilioideae. Of the genera included under the subfamily 

 Lilioideae (Krause, 1930), Lilium, Fritillaria, Tulipa, Lloydia, and 

 Erythronium have been shown to have a tetrasporic embryo sac, 

 usually of the Fritillaria type; Nomocharis is uninvestigated. Only 

 Calochortus has been found to have an embryo sac of the monosporic 

 eight-nucleate type (Cave, 1941). It is interesting to note that this 

 genus also differs from the remaining members of the Lilioideae in 

 some other respects. The fruit of Calochortus is a septicidal capsule, 

 while that of the other genera is loculicidal. In Calochortus the 

 chromosome numbers are 6, 7, 8, 9 and 10 (Newton, 1926; Beal, 

 1939; Owenby, 1940), while in the remaining genera the number is 

 usually 12. On the basis of these and some other differences in the 

 structure and germination of the seed, Buxbaum (1937a, o) proposed 

 that Calochortus should be transferred to an independent subfamily 

 under the Liliaceae. This opinion, which is strongly supported by 

 the embryological data brought forward by Cave (1941), may now 

 be accepted without reservation. 8 



Conclusion. There are several other notable examples of the aid 

 which embryology has rendered in the solution of taxonomical 

 problems, but it is unnecessary to cite all of them here. What 

 has already been written is sufficient evidence of the value of em- 

 bryological data" in an elucidation of the interrelationships of fam- 

 ilies and genera. Although it is not claimed that these data will 

 always prove important, they should form a part of any thorough 

 taxonomic analysis. It is worth while, therefore, for every student 

 of embryology to try to assess the bearing of his observations and 

 those of his predecessors on the taxonomic position of the group of 

 plants which he is studying. Many of his conclusions are bound to 



"See also Schnarf (1948). 



