CYTOGENETICS AND EVOLUTION OF THE GRASS FAMILY 1 79 



Second, there are two types of starch grains in the grasses, compound and 

 simple. Although there are some exceptions, the tribes and subfamilies are 

 generally uniform in the type of starch grain which they possess. Compound 

 starch grains are characteristic of most Festucoideae, the Chloridoid-Eragros- 

 toid complex, the Arundineae, Oryzeae, and probably Bambuseae. Simple 

 starch grains are found almost throughout the Panicoideae, as well as in the 

 Hordeae, Brae hyp odium, and Bromus among the Festucoideae. 



Third, there has been marked differentiation in respect to the cells sur- 

 rounding the vascular bundles of the leaves (W. V. Brown). This is 

 relatively little developed in the Bambusoid, Danthonoid, and Festucoid 

 groups, but in both the Panicoideae and the Chloridoid-Eragrostoid group, 

 as well as in some of the smaller groups, there has been extreme specializa- 

 tion in respect to both the parenchyma sheath cells and the chloroplasts which 

 they contain. The connections between these anatomical features and the 

 functions of photosynthesis, food storage, and food transport in the leaves 

 have not yet been established, but certainly deserve attention. 



Finally recent work (Al-Aish, 1956) has shown that grasses differ greatly 

 in their response to a weed-killing chemical, isopropyl-iV-phenyl carbamate 

 (IPC). The Festucoideae, Danthonieae, and Stipeae are all easily killed as 

 germinating seedlings by weak doses of this chemical, while the other groups 

 tested, Panicoideae, Chloridoid-Eragrostoid group, Oryzeae, Aristida, and 

 Streptochaeta, are strongly resistant to it, to such an extent that in weak 

 doses IPC may stimulate germination. This difference is associated with 

 the ability of the IPC-resistant species to carry on respiration in nearly 

 anaerobic conditions. These latter species, furthermore, are mostly tropical 

 or subtropical in distribution, while the grasses which are susceptible to IPC 

 occur exclusively in temperate regions. Here is a connection between physio- 

 logical divergence and geographic distribution, which, when further analyzed, 

 may throw much additional light on the evolution of the grass family. 



This discussion of the new classification of grasses and its bearing on evo- 

 lution leads to the following conclusions. The new system of classification 

 will not lighten the burden of the taxonomists who are charged with the task 

 of identifying and naming the thousands of grass specimens which are col- 

 lected each year. In fact, the wiser course for such work may be to retain 

 the artificial keys constructed on the basis of the traditional system, since 

 they are known to work reasonably well, and many of the more newly dis- 

 covered diagnostic characteristics are rather hard to determine on the basis 

 of the usual dried specimens. On the other hand, recognition of the newer 

 characteristics and their bearing on evolutionary relationships will undoubt- 

 edly lead to a far better understanding of the nature and direction of evolu- 

 tion in this remarkable family. Furthermore, it has practical significance. It 

 can give the grass breeder a clearer idea of the extent of the "gene pool" 

 from which he can "borrow" genetic material by hybridization, with which 



