178 NATURAL SCIENCE. march. 



could occupy, and in every case the temporary and local climate 

 must be indicated by the local flora, while the succession in any 

 one place may be relied on as holding good over a very extensive 

 area. 



Looking at the Palaeozoic plants a little more in detail, we 

 remember that coniferous and taxine trees grow now in very different 

 latitudes and climates. There is, therefore, nothing so very 

 remarkable in their occurrence. The great group of Cordaites may 

 have been equally hardy, but it is noteworthy that its geographical 

 distribution is more limited. In Europe, for example, the genus is 

 much more characteristic in France than in Britain, perhaps in 

 •connection with the warmer climate to the southward. Ferns and 

 lycopods and mare's-tails are also cosmopolitan ; but the larger 

 species now belong to the warmer climates, and nowhere in the 

 present day do they become so woody and so complex in structure as 

 in the older geological periods. 



The natural inference would be that in the old coal period the 

 geographical and other conditions must have conspired to give a some- 

 what uniform and moist climate over a great part of the earth's 

 surface. The geographical arrangements, so far as known, indicate 

 this, and the distribution of animals points in the same direction. In 

 America, for example, the great eastern and western ranges of 

 mountains were yet in embryo, and a large part of the continent was 

 occupied with shallow water or with swampy plains scarcely above 

 the sea-level. The batrachians and insects of the land and the varied 

 forms of animal life in the sea alike point to a climate at least mild 

 and uniform. We must also take into account the probability that 

 there was a larger amount of carbon-dioxide in the atmosphere than 

 at present, which would have greatly impeded radiation from the 

 ground, while the moisture exhaled from the vast swamps and 

 morasses of the period would have produced a similar effect. 



It would, however, be a mistake to suppose that there were not 

 local differences of climate. I have in my " Acadian Geology " advo- 

 cated the theory that the great ridge of conglomerate on the northern 

 border of the coal-field of Pictou, in Nova Scotia, ma}'' have been an 

 ice-formed ridge on the margin of the deep morass in which the thirty- 

 six feet seam was deposited. In this case a sea occasionally ice-laden 

 may have approached within a short distance of forests of Sigillariae 

 and Lepidodendra, and this in the middle of the coal period. On the 

 whole, therefore, we should postulate for the Palaeozoic flora not so 

 much a high temperature as uniformity and moisture. This seems 

 also to accord v/ith the prevalent character of the foliage and the 

 structures of the remarkable acrogenous trees of this period. 



As to the early Mesozoic flora, the indications are that it was an 

 invader from the Southern Hemisphere, for which the intervening 

 Permian period had prepared the way by destroying the preceding 

 Palaeozoic forests. This was probabJy effected through the agency of 



