mj 



to Brazil, he refers the two species also occurring in the West Indies: 

 Conocarpus erecta L. and Anona palustris L. I have in my earlier 

 paper, though with some doubt, included the two last-mentioned 

 species in mentioning the mangrove formation, as they commonly 

 occur among the Laguncularias inside the mangrove vegetation. 

 According to my last observations I consider it more natural to 

 keep them apart from the true mangrove formation, and to refer 

 them, along with some other species, to a special formation, to 

 which I shall return later on. 



While the conditions of the level are of great consideration 



Kig. 2. Saltriver Bay on the north side of St. Croix. (F. B. phot.). 



for the placing of the species belonging to the mangrove vegeta- 

 tion, I cannot, in any case with respect to the Danish West 

 Indies, agree with Schmidt (1. c. p. 8) when he writes that it is not 

 especially the consistency of the soil, but the conditions of the 

 level, which mark the boundary between the mangrove and the 

 sandy shore vegetation. In the West Indies it is firstly, more or 

 less, the exposure of the locality which is the deciding factor as we 

 never find any of the representatives of the mangrove on the ex- 

 posed, but only on the sheltered coasts. And as on the exposed 

 coasts, the rocky shores excluded, we have only the white coral- 

 sand beach where the sand vegetation grows, and |only muddy soil 

 on sheltered coasts, the consistency of the soil will in any case 



