432 A CENTURY OF PROGRESS IN THE NATURAL SCIENCES 



Schnarf, Copeland, Maheshwari, Battaglia). Unfortunately for purposes of 

 classification and evolutionary interpretation, the distribution of embryo-sac 

 types appears to be largely haphazard, although Fagerlind (1944) supposes 

 tetrasporic gametophytes to be indicative of relationships within at least some 

 taxa. The shining exception is the restriction of the Oenothera type (mono- 

 sporic, four-nucleate) to Onagraceae; this phenomenon permits expulsion from 

 the family of the anomalous genus Trapa, which has an eight-nucleate embryo 

 sac and other structural peculiarities. The genus Calochortus, according to Cave 

 (1941), should be excluded from Tulipae and Lilioideae because of its lack of a 

 Fritillaria-ty-pe embryo sac and certain characteristics in chromosome number. 

 Suessenguth (1921) and Salisbury found in embryo-sac features a further verifi- 

 cation of a connection between Ranales and Alismatales; the importance of a 

 multinucleate embryo sac as indicating relationship between Piperales and 

 Arales or Pandanaceae, is denied by Maheshwari (1950). 



The explanation of "double fertilization" and the homologies of the polyploid 

 endosperm have generated a great deal of discussion and widely different opin- 

 ions. The occurrence of a "double fertilization," together with the general uni- 

 formity of other features of the ovule and particularly the embryo sac, has been 

 given the greatest importance as providing conclusive evidence that angiosperms 

 have had a single basis (Sargant, Schnarf, Fagerlind, Whitehouse; Parkin, 

 1952). However, Battaglia believes that the antipodals and the polar (proendo- 

 spermatic) cell are actually latent primary-endosperm initials awaiting physio- 

 logical stimulus, and that "double fertilization" is a misnomer for the necessary 

 stimulation furnished by mitosis. "PolyanUpody, in the phylogeny of angio- 

 sperms, should, therefore, he an indication of primitiveness" (1951, p. 96). 

 Brink and Cooper (1940) have stressed that "double fertilization" gives the 

 embryo the advantage of a rapidly developing food supply through polyploid 

 heterosis. Glisic (1929) used features of the endosperm and haustorial similari- 

 ties to relate Orobanchaceae closely to Scrophulariaceae, and Campbell (1930a) 

 shifted Paulownia from Scrophulariaceae to Bignoniaceae despite differences in 

 endosperm. 



C. Male gametophyte: The male gametophyte of angiosperms is the germi- 

 nated pollen grain with its fully developed pollen tube. Although it is gener- 

 ally assumed to have been derived, with a further reduction of prothallial tissue, 

 from some ancestral gymnosperm, we know very little about either its origin or 

 its subsequent evolution (Maheshwari, 1949, 1950; Battaglia). Because of the 

 important role now being played by microfossils in the comparative dating of 

 phytogeographical, climatological, and archeological events, there is currently a 

 great deal of interest in pollen and palynology. The importance of fresh pollen 

 in allergy, also, has led to the accumulation of the kind of comparative data re- 

 quired for systematic purposes. The features of the stamen and the male gameto- 

 phyte commonly accepted as of classificatory value include: (1) the nature of 

 the anther tapetum, whether glandular or plasmodial; (2) the mode of division 

 (simultaneous or successive) of the microspore mother cells and the resultant 

 configuration of the spore tetrads; and (3) the morphology of the pollen grain, 

 comprising particularly size, shape, and symmetry, the number and position 

 of apertures (germ pores) and furrows (colpae), the adornment of the outer 

 spore wall (exine, sporoderm), the number of nuclei in the pollen grain at the 



