362 M. Brongniart on the Vegetation of 



Under the same latitude, this proportion becomes much great- 

 er in the islands. Thus, in the West Indies the proportion of 

 Ferns to the Phanerogamic plants appears to be nearly as one to 

 ten, in place of one to twenty, which is that of the most favoura- 

 ble parts of the American continent. In the South Sea Islands, 

 this proportion, in place of being as one to twenty-six, as in the 

 continent of India, and the tropical parts of New Holland, be- 

 comes as one to four, or as one to three. At St Helena and 

 Tristan d'Acuna, the proportion of these classes of vegetables is 

 as two to three. Lastly, in the Island of Ascension, considering 

 only the plants evidently indigenous, there appears to be an 

 equality between the Phanerogamic and the vascular Cryptoga- 

 mic plants. It is seen from these examples, that the smaller the 

 islands are, and the more remote from the continents, the great- 

 er does the proportion become which the Ferns, and allied fa- 

 milies, bear to the total number of the other vegetables ; and it 

 may be conceived, that, were islands, like those just mentioned, 

 alone to exist in the midst of a vast sea, where they would only 

 form scattered points, or small archipelagoes, without any great 

 continent, the proportion of Ferns would probably be still 

 greater ; and, in place of an equality in the two great classes of 

 vegetables which we compare, we might see the vascular Cryp- 

 togamic plants greatly predominating over the Phanerogamic. 

 This is what took place at the epoch of the formation of the 

 coal deposits, and these considerations of botanical geography 

 should of themselves incline us to think that the vegetables 

 which gave rise to these deposits grew upon archipelagoes of 

 islands of small extent, at an epoch when no great continent 

 rose above the surface of the waters. 



The disposition of the coal deposits in interrupted lines, which 

 have been called basins, and, compared with lake successions or 

 valleys, has at least a considerable analogy to the disposition of the 

 islands which, representing the ridges of submarine mountain- 

 chains, are generally placed in rows. Lasdy, the interruptions of 

 the coal formation, and, on the contrary, the great extent and con- 

 tinuity over large spaces of the transition limestone formations, 

 which may be considered as the deposits formed in the sea which 

 surrounded these islands, seem to confirm this hypothesis. 



Count Sternberg and M. Boue, proceeding solely on geological 



