92 
W. Watson. 
condition, the flora being similar to that of peat mosses in other 
localities. This is the only area of Somerset in which Hypnum 
gigantenm is known to occur; on the other hand the following plants 
which occur in similar associations in other parts of Somerset are 
not present or have not been noticed :— Peplis portula, Ranunculus 
flammula (a small form common in the bogs on the Quantocks), 
Wahlenbergia hederacea, Polytrichum commune , Hypnum jluitans. 
It is also to be noted that many of the plants found on the heath 
are now absent from the Somerset lowlands and are only found in 
such hilly regions as Exmoor and the Quantocks. Some bryophytes 
(Lepidozia setacea, Odontoschisma sphagni, Leptoscyphus anomalus, 
L. taylori) characteristic of peat mosses have not as yet been found 
in any part of Somerset. The wetter pockets of Sphagnum recurvum 
represent a more primitive condition of the heath than those of 5. 
cymbifolium, the Sphagnetum cymbifolii being a later development. 
(2) Sphagnetum subnitentis. As the heath became drier 
Sphagnum subnitens became dominant. The Hypna and Sphagna 
which are now prevalent in the lower swampy portion and in the 
pockets became less abundant and gradually retreated to the wetter 
areas. A slight erosion of the humus from these areas and its 
retention and increase on the drained areas would then enable the 
next stage of succession to occur. 
(3) Molinietum. In many cases of progressive desiccation of 
a peat moss, the Sphagnetum does not directly give rise to a 
Molinietum but to an intermediate cotton-grass association (Erio- 
phoretum) occurs. 1 There is no certain evidence of this succession 
stage occurring in this area. Eriophorum augustifolium is doubt¬ 
fully recorded for the district in Murray’s “ Flora of Somerset,” but 
though some parts of the heath seem to be suitable habitats, it has 
not been noticed either by myself or Rev. E. S. Marshall. 
After the invasion of the Molinia the drainage areas became 
more constant and tussocks of this grass were formed. The increase 
in height of these tussocks has been chiefly due to the decay of the 
Molinia and Myrica. The furrows have been undoubtedly eroded 
but their ultimate increase in depth by such erosion is slight. 
During the wettest portion of the year (December is often the 
wettest month in this district) when erosion of an unprotected 
surface would be at a maximum, there is a superficial layer of 
dying vegetation, and the erosion does little more than compensate 
for the additional humus formed. Nardus stricta does not seem to 
1 VV. G. Travis, “ Plant Associations of some Lancashire Peat mosses.” 
