J. Starlde Gardner — Development of Dicotyledons. 163 



Mijrlca, large-lobed leaves referred to Acer, and lai'ge simple leaves 

 referred to Ficus aud others, together with leaves of an immense 

 Fan Palm. None of the determinations seem well grounded except 

 Myrica, but there are several leguminous fruits, antl the flora, as a 

 whole, is easily and immediately recognizable, and would suffice to 

 define the age of any beds in which it occurred. 



Following in order is the Bournemouth flora, of great variety and 

 extensively collected. It is very unlike any of the preceding, and 

 were its age not absolutely known beyond the faintest possibility of 

 doubt, it would have been referred to the Miocene, though actually 

 older than the Bracklesham beds. It greatly resembles the Eocene 

 floras from the South of France and of the Lignitic Series of 

 America. Its affinity to the Oligocene floras of the rest of Europe 

 appears so great that it is doubtful whether any distinct line can be 

 drawn between them, but a little higher at Hordle, some of the 

 Older Eocene Reading types seem to mingle with it. 



All these Eocene floras are in beautiful preservation, and their 

 stratigraphical position is definitely ascertained. A careful study of 

 them will be repaid by much valuable information regarding the 

 order of appearance of the genera of Dicotyledons, and would enable 

 us to at once recognize the ages of any Eocene floras of adjoining 

 countries. We should be able to argue from the known to the un- 

 known, instead of the reverse as heretofore practised, and to con- 

 struct for the first time a definite chronological sequence with which, 

 other floras might be contrasted and compared. The series can only 

 be paralleled by the fossil floras of South-Eastern France, so beauti- 

 fully described by Saporta. 



In utilizing them for comparison with the fossil floras of other 

 counti'ies, the differences of latitude and longitude must be taken 

 into account. Nor have we any right to suppose that all the plants 

 preserved from an immense number of localities grew at the same 

 elevation above the sea, while they may also have lived on very 

 different stations, and under relatively dry or moist climates. Any 

 of these influences, singly or combined, might profoundly modify 

 floras, which, though perfectly contemporaneous, would then present 

 very different compositions and aspects. Such considerations are so 

 very obvious, that it svould seem almost unnecessary to insist on 

 them, yet we find in examining works on the Miocene floras that 

 they have not been regarded. Floras from Spitzbergen in the North 

 to Australia in the South have been classed as Miocene from a very 

 slender fancied resemblance to those of Switzerland, and great series 

 of strata have been assigned without sufficient reason to that age. 

 not only in Central Europe, but in such distant lands as Greece, 

 Madeira, Borneo and Sumatra, Sachalin and Alaska, and in fact 

 wherever other evidence of age was absent. The resemblance of 

 some of these floras to those of Switzerland is in many cases very 

 shadowy and quite insufficient to justify any conclusion as to their 

 age ; but it is more than doubtful whether many of the supposed 

 Lower Miocene Floras of Switzerland classed as Aquitanian are not 

 themselves as old as the Eocene, and others, as the Mayencian, Oligo- 



