450 SKETCH OF PALEOBOTANY. 



per cent), and very much larger in the Tertiary than in the living flora 

 (14 per cent), while the Gamopetalse, which constitute only 5 per cent 

 in the Cretaceous, reach 15 per cent in the Miocene, and 46 per cent in 

 the living flora, here exceeding the Polypetalse. From these facts it is 

 evident that the order of development is such as I have here given it, 

 and that the type of the future is to be not the Polypetalse but the 

 Gamopetalae. These conclusions are independently corroborated by a 

 large mass of evidence of other kinds, but space forbids me to adduce 

 it in detail. I may simply say, however, that just as the closed ovary 

 of the Angiosperm in general furnished a condition for the development 

 of that class at the expense of the unprotected Gymnosperm, so the two 

 tloral envelopes of the Polypetalae and Gamopetalse enabled those divi- 

 sions to outstrip the Apetalse with its single floral envelope ; and since 

 this advantage is proportional to the degree of protection secured, the 

 Gamopetalae, with their tubular corollas are manifestly better adapted 

 to survive in this respect than the Polypetalae. This is the chief argu- 

 ment, and, putting it with that from paleontology, it seems sufliciently 

 conclusive without detailefd support. 



Discussion of Diagram Ifo. JJ.— In this diagram the time equivalents 

 are the same as in the last, but only the more important types are rep- 

 resented. The Ehizocarpeae, Ligulatae, and Gnetacese are oniitted, and 

 the Dicotyledons as a whole are shown, disregarding their subdivision 

 into Apetalse, Polypetalse, and Gamopetalae. A figure is added represent- 

 ing the total of all the formations, and this is probably the most impor- 

 tant of them all, as least aifected by the gaps and fluctuations in the 

 record. No account could, of course, be taken of the living flora, as is 

 done in Diagram No. I, for while between the fossil and the living-floras 

 there is a similaritj' in the pi-oportion that the types in each bear to the 

 sums of such floras, no such analogy holds between the number of species 

 actually known in any fossil flora and the number in the living flora. This, 

 at least, is true of the total floras and of all the types except, perhaps, 

 the Cycadaceai and the Couiferte. But even here the comparison would 

 fail to express the rapid decline which these for ms have evidently un- 

 dergone, at least so far as tbe number of their species, which represents 

 their diversity, is concerned. 



While the diagram is of little service as a means of representing the 

 true development of each type of vegetation or of the general flora of 

 past ages, it has considerable value as an exponent of the true charac- 

 ter of the phyto-geologic record. It shows more clearly and more strik- 

 ingly than any words or figures could do the great differences that 

 characterize the different periods of geologic time in their susceptibility 

 to deposit, preserve, and afterwards exx^ose to scientific investigation 

 the vegetable forms that constituted the floras of those periods. While 

 this is well shown for the several dominant types it is especially obvi- 

 ous in the figure illustrating the entire flora. Here are brought promi- 

 nently into view, first, the age of island vegetation in the Carbon- 



