80 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



usually found along slow-moving streams in typically swampy 

 areas. Hut there are many such areas on the coastal plain of 

 Long Island where the species might be expected, notably near 

 Babylon (16) and along the Connecticut (Carman's) River from 

 Yaphank to South Haven. But it has never been found at these 

 and several other likely localities. Just what the factors con- 

 trolling the appearance of the species on Long Island are, it is 

 impossible to say. 



The swamp at Merrick stretches from about three quarters of a 

 mile north of the Long Island Railroad, along the banks of a 

 stream to the salt marshes, which are about three quarters of a 

 mile south of the railroad. The best present development of the 

 swamp is at the northern end, on the property of Harold Bunker, 

 Esq., who has carefully protected the existing grove. A good- 

 sized pond, caused by the damming of the stream about a hundred 

 years ago, divides this northern part of the cedar swamp. All 

 of the region now occupied by the pond was once covered by white 

 cedar trees, submerged stumps of which may still be seen in the 

 clear water. Plate 6 shows the present general aspect of the 

 region, and plates 7 and 8 the upper ends of the pond and 

 the growth of cedars beyond. These trees mark the northern 

 limit of the grove. To the southward the swamp has suffered 

 much from fire and from the pumping operations of the City, a 

 reservoir having been made just north of the railroad, thereby 

 clearing the trees from this area. To the south of the railroad 

 the cedar trees stretch uninterruptedly to the salt marshes near 

 which there is a conspicuous fringe of dying and dead trees. 



One of the most characteristic plants of the undergrowth in 

 white-cedar swamps is Rhododendron maximum (19, 22), and, 

 in the south, Alagnolia. Neither of these are found in the swamp 

 at Merrick, and indeed the character of the undergrowth there is 

 such that several interesting problems are suggested by a study 

 of the floristic content of the whole swamp. 



The commonest tree, besides the Chamaecyparis, is Acer caro- 

 linianum, if this be not a mere form of A . ruhriim. More scattered, 

 and much less common are Nyssa syJvalica, Qiiercus rubra, and 

 strangely enough. Sassafras, which is ^•e^^• plentiful in the dr}' 

 reaches of the adjacent coastal plain scriil). but seems perfectly 

 at home also in the deep shade of this moist forest. See plate 

 9 for general asi:)ect of the interior of the swamp. 



