42 THE RELATION OF PLANTS TO TIDE-LEVELS 
portions of the Spartina turf that are hanging down into the stream. Along 
all the fresh-water rivulets shown on the map between 0 and 200 north the 
Spartina keeps upon the firm peat at the side, 6 or 8 inches above the constantly 
flooded gravel bottom of the rivulet. The rhizomes and roots must often reach 
down nearly to the fresh-water level. Only at one place about the harbor has 
S. glabra been found growing where its rhizomes are constantly immersed. 
This was in a pool of 10 feet in diameter filled by fresh water falling from the 
flume south of the mill (plate tvs). The surface of this pool was at the 5-foot 
level, and the Spartina is rooted in soil some inches below this, so that its roots 
and rhizomes were surrounded by fresh water for 6 or 7 hours each tide. In 
1910 the water was shut off from this millrace and it is now dry. The only 
seed plant growing in the beds of these fresh-water rivulets is Lileopsts lineata, 
which we have already mentioned as the only other seed plant characteristic 
of the Spartina belt. In the bed of one of these rivulets near 150 north, 
between the 4 and 6 foot levels, hundreds of these tiny plants with their bladeless 
leaves lie appressed to the gravel, while the cold fresh water runs over them 
for 6 to 9 hours each tide. 
The steepness of the shore, the character of the soil, and the abundance of 
the supply of fresh water are very different on different parts of this eastern 
side of the harbor. In evident consequence of this we find that the upper 
limits reached by Spartina glabra, and by the other plants with which it comes 
into competition on the upper portions of the mid-littoral belt, are much more 
variable than on the Spit, where we found a very regular upper margin of the 
dense Spartina running along the 6.5-foot tide-line. For example, at 200 
north there is a rather well-drained bit of mid-littoral beach. In this region the 
pure dense growth of Spartina ceases at the 5.5-foot level. At the 6-foot level 
it becomes rather equally mixed with S. patens, which becomes dominant from 
_this level up to 7.5 feet. Only a few scattered small plants of S. glabra reach 
to the 6.5 and %-foot levels. At 350 north, where the east shore between 5 and 
7.5 feet is of coarse gravel, and supplied with very little fresh water trickling 
over the beach at low tide, the pure stand of S. glabra ends at 5.5 feet, where it 
becomes mixed with Scirpus americana and with still fewer plants of Spartina 
patens. Above the 6-foot level the latter becomes dominant and S. glabra is 
sprinkled more and more sparsely with it, and with Scirpus americanus, up to 
the 7-foot level. Above this the S. glabra is wanting altogether. 
From 400 to 500 north, where, in some places fresh water is trickling over 
the beach from above the 8-foot level, and in other places where it is seeping 
out of the gravel at lower levels, we find the dense S. glabra going up to the 
6.5-foot level, in places where the turfs are kept continually moist by fresh 
water running constantly about them. Here, as elsewhere, it does not grow at 
all where fresh water is running directly over the soil in which its rhizomes are 
embedded. Usually this is evidently because the fresh-water rivulet cuts away 
the peaty soil down to the underlying gravel in forming its little channel. The 
rhizomes, except in the pool below the mill-wheel, are always in soil high enough 
above the beds of these rivulets so that the soil-water always remains more or 
less saline even at low tide. It is probable also that the water constantly 
running over the beach may keep the air about the leaves of the short Spartina 
of the upper levels more than usually moist during low tide and thus, by 
reducing transpiration, enable it to creep a little higher up the beach. Along 
