718 THE RELATION OF PLANTS TO TIDE-LEVELS 
erass that keep it down to 6.5 feet and so determine the lower limit to which 
Spartina patens usually reaches. The upper limit of S. patens is probably 
determined by the physical character and salt-water content of the soil, both 
directly and by the competitors favored or excluded by these conditions. 
The lateral boundaries of the patches of Spartvna on the shore seem to be 
determined by the local characters of the soil. Thus, for example, gravelly 
sections of this upper littoral beach are pretty sure to be destitute of this grass, 
which is there replaced by Salicornia, Sueda, or Atriplex. Of course, it is 
possible that the coarser soil may be such in consequence of the lack of the 
Spartina as a binder. Wherever a fresh-water rivulet trickles across the upper 
littoral beach the band of 8. patens is broken and Scirpus americanus pushes in 
to occupy the soil saturated with fresh water. Even Spartina glabra may, in 
these moist places, push above its usual upper limit, on knobs of peat that are 
high enough to avoid being constantly wet with the fresh water at low tide. 
Juncus Gerardi: In the upper littoral belt this rush is found in dense turfs 
over very considerable areas of the Marsh (plate xx a), chiefly between the 
?-foot and 8-foot levels, though sometimes higher, as will be seen from the work 
of Professor Conard (plates x1, xx1, and xx11). Elsewhere about the harbor 
only two patches of it are found, and these are small in area and the shoots of 
Juncus are intermingled with those of other species, oftenest with those of 
Spartina patens. In one of these areas (200 north by 1,060 east at 7.5 to 8.25 
feet), the patch of Juncus is 0.5 meter wide and 4 meters long. There is here a 
sparse admixture of Solidago sempervirens and Scirpus americanus. On the 
west shore (1,700 north at the 8-foot level) there is a dense turf of this Juncus, 
of 2 square meters area, the only one on the whole west side. On the Spit this 
rush has not been seen at all. 
The distribution of this Juncus can evidently be studied best on the Marsh 
and will therefore be left for Professor Conard to discuss. We may simply 
remark at this point that the deep, peaty soil inhabited by this species is practi- 
- cally wanting, except on the Marsh. (See plates x1, xx1, and xxu1, and fig. 3, 
p. 111.) It is also interesting to note that at the one point on the west side 
where this rush occurs it is accompanied by the two plants usually associated 
with it on the Marsh, Distichlis and Spartina patens. The soil on which these 
three species here find congenial conditions is a deep, peaty muck like that of the 
Marsh, which is formed chiefiy by the sedimentary deposits from the very con- 
siderable stream that enters this side of the harbor at 1,650 north. 
Sueda maritvma in the upper littoral belt: This low, glaucous annual is 
distributed abundantly along the south shore of the Spit (plates Iv 4, v, viz B, 
and xiv), and occurs in rather frequent smaller patches on well-drained portions 
of the Marsh, but is rarely found on the east or west sides. On the Spit Sueda 
is distributed pretty generally from end to end, chiefly between the 6.5 and 7.5 
foot levels, though it occasionally gets down to the 6.25 or up to 8 foot levels. 
In some stretches of the upper littoral beach it is the dominant species in the 
belt immediately above the Spartina glabra. On other parts of this beach Sueda 
may be crowded downward into the S. glabra or upward toward the 8-foot level, — 
or occasionally be crowded out altogether, by such competitors as Spartina 
patens, Salicornia europea, or Distichlis spicata, which are the other species that 
may become dominant in this belt. In still other portions of this belt Sueda 
may occur as a mere sprinkling over areas dominated by one of the three species 
