86 THE RELATION OF PLANTS TO TIDE-LEVELS 
west shore (800 to 930 north, 1,600 north), on the east side on the pier (950 
north), and finally on several small areas of the Marsh (e. g., 0 south by 1,200 
east). In other areas where Distichlis is still dominant, there may be a large 
admixture of Sueda or Salicornia, or even a scattering of Atriplex arenaria 
(e. g., 2,700 north by 0 east, 2,850 north by 800 east). 
The soil on which Distichlis grows is a partially drained, peaty muck, occa- 
sionally mixed with sand, as, for example, at the western end of the Spit. More 
rarely this grass is found in comparatively fine sand, in which on the Spit (750 
east) it gets up above the 8-foot level. Nowhere about this harbor was Distichlis 
found growing in the shade or in soil wet with fresh water. The latter fact may, 
of course, indicate that the fresh water is directly injurious, or perhaps we 
should say that the semidiurnal alternation of fresh and salt water is unendur- 
able. Or on the other hand, its absence from wet areas may mean that it meets 
other competitors, or meets some of its usual competitors at a greater disadvan- 
tage on wet soils. ‘This grass lives on soils of quite varied character, taking all 
sides of the harbor into account, which indicates that it could occupy many 
other areas than at present if it were not for its competitors. In fact, then, it 
seems clear that the chief influence determining the limits of the Distichlis in a 
horizontal direction along the beach is the competition of its neighbors. 
It is difficult to discover what external condition fixes the lower limit of 
Distichlis at 6.5 feet. It may be that the shoot will not endure a submergence 
of more than 2 or 3 hours each tide or that the rhizome can not withstand sub- 
mergence longer than 3 or 4 hours per tide. On the other hand, it may well be 
that it can not compete successfully with Spartina glabra, with which it is 
nearly always in contact at its lower margin. The upper limit of this grass is 
quite variable, but it has proven impossible to determine whether this is fixed 
directly by some character of the soil, such as the water-content, or by the com- 
peting plants there present. 
It is, of course, realized, as has been mentioned in other cases, that the 
particular conditions which limit the distribution of any species can not be 
determined by field observation alone, but that resort to experimental study will 
be necessary to do this with certainty. At the start, however, field observation 
must be relied on to indicate the possible factors from among which the experi- 
menter may hope to select the actual controlling factor or factors. The sug- 
gestions offered in connection with most species in this paper are not made in 
the belief that they are finally established as causes of the distribution found, 
but as suggestions likely to prove useful to the experimental investigator. 
The distribution of the nine other species of angiosperms found in this belt, 
but which seldom dominate any area above a few square decimeters in extent, 
may now be briefly indicated, taking them up in alphabetical order. 
Atriplex arenaria: This hoary-leafed annual is found very sparsely scattered 
on the eastern half of the Spit (plates v and xiv), on the Marsh (plates xz and 
XII1), and on the old pier at 950 north on the east shore. Its usual range is from 
the 7-foot to the 8-foot levels, though its extreme range is from 6.25 to 8.75 feet. . 
It occurs in rather sandy soil on the Spit, but is most frequent in the turf of 
Spartina patens along the edges of the tide-streams and. ditches on the Marsh. 
Atriplex patula hastata is, in most years, very much more abundant than A. 
arenaria. It is also more widely distributed about the harbor and has a greater 
vertical range (plates v, x1, and x11). This species has been recorded on the 
