94 
Nor did T succeed in finding this plant upon Green Mountain in 
Mt. Desert Island, though a large part of the bare ledges of its sum- 
mit and of its western flanks was searched on the 7th of August of 
this year. Yet in so large an area it might easily elude discovery, 
and, even if now absent, it may once have existed previously to the 
fires which have so repeatedly overrun this mountain, and which 
have evidently made great changes in the character of the vegetation. 
On the rocks just east of the observatory I noticed limited quantities 
of Hudsonia ericoidcs,^V\c\i may possibly have been mistaken for 
Corema. Nor. did I observe Empetrum on that mountain, though so 
abundant on the headlands near the shore. 
I can, however, announce the discovery of Corema in the southern 
part of Mt. Desert Island, near Seal Harbor, in three distinct locali- 
ties. These are all upon the ridge called Jordan's Hills on the Coast 
Survey Chart, and on that portion of it which has been named Barr's 
Hill. My attention was called to the first of these localities by Mr. 
Arthur Chase of Haverhill, Mass. It is upon the western brow of 
the hill, upon a broad rock terrace overlooking Long Pond, about 
150 feet above the level of the pond. Within a space of 50 feet 
square occur eight or ten patches of Corema, in scanty soil upon 
almost bare rock, disputing possession with Gaylussacia resinosa and 
Vacclnium Pennsylvaniaim. Its other neighbors were Potentilla tn- 
dcntdta, Kalmia angustifolia and Viburmwi nudum. The nearest 
trees were 40 or 50 feet distant, and were mainly Fopuliis alba,_ van 
populifolia, and Pinus Sirobus. The rock is a sienite containing 
quartz and felspar in about equal proportions, with a very small 
admixture of hornblende, and the comminuted fragments make a 
sand of similar composition. A few days later, a second locality was 
found by Rev. Archibald M. Morrison of Orange, N. J. This is 
upon the eastern side of the ridge, a little northward from the head 
of the old path from the Seaside House, and nearly 300 feet above 
tide. Here there is but a single patch of the plant, six or eight feet 
in diameter, and here too it seems likely to be eventually stifled by a 
growth similar to that already mentioned. The rock is the same, 
with deeper soil. On the 19th of September, I found a third locality 
upon the southwest brow of the ridge, northeast from the *' sea-wall 
of Long Pond, 100 or 120 feet above tide. Here four or five patches 
are found under surroundings much like those of the other localities. 
The underlying rock is the same. The indications of dying out are more 
apparent here, there being a large amount of the dead planton tne 
margins of the patches, and one large patch of several faet in dianiete 
was entirely dead, except for a few inches in the centre. We may 
note that, in all these localities upon Barr's Hill, the soil is less 
exclusively silicious than in any previously seen, and that the P^^^' 
lent pine is not Pinus rigidus, but Finns Sirobus. This hill contain 
hundreds of acres of bare, horizontal ledges of rock just as w 
adapted to the growth of Corema as is the limited space ^<r^"^ f.^e 
cupied by it, and yet repeated and careful examination of its sur ^^ 
by numerous observers have brought to light not enough ot 1 
cover a tenth of an acre. ' ^^^^ 
The study of the various localities of Corona which I nave 
1 
