PLANTS OF NORTHERN GUJARAT. 
233 
Of these, Andropogon contortus is useless as fodder, because cuttings 
are made too late. Andropogon Sc/iosnanthus is too strongly scented to be 
of much value, but it is fed to buffaloes. Aplvda varia is a weak grass and 
does not form colonies so freely as the others. Andropogon foveolatus is 
rather patchy. The remainder constitute the bulk of the finest bales of 
cut grass. Anthistiria ciliata, which is probably the best fodder in 
India, is not so universally prevalent here as it is south of the Tapti. 
Locally, the “ jhinjawa ” ( Andropogon annulatus ) is preferred to all 
other kinds. It is an extremely slow grower, and matures in the early 
winter. Consequently, while the monsoon cuttings contain chiefly 
Ischcemum rugosum, Iseilema Wightii and Anthistria ciiiata, the winter 
cuttings contain mainly “ jhinjawa ” ( Andropogon annulatus ). 
The habitats of these grasses differ. While the bulk are xerophytic, 
Ischcemum rugosum and Iseilema Wightii prefer the wetter and more 
low-lying portions of the “ birs ” and especially the places where the 
ground has given way in waterholes. So regular is this characteristic, 
that when riding over “ birs % * when the grass is up, a horseman is 
almost safe from the fear of landing his horse in a hole if he threads 
his way among the patches of “ jhinjawa ”, easily distinguishable by 
its paler colour and finer stems. 
Andropogon Schcenanthus is a particularly xerophytic species, both 
physically and physiologically. It matures very early and with the 
minimum of rainfall, and thereafter maintains itself without appreciable 
dryage for months. No doubt the oily products which it contains act 
as reserve material and maintain turgidity. This grass can be seen 
along the banks of the Bombay, Baroda and Central India Railway on 
the Ajmer line a little north of Ahmedabad. 
These grasslands are nowhere treeless. Besides the trees mentioned 
under Xerophytic Woodland, and Butea Savannah, we have noted 
Capparis grandis , Morviga pterygosperma, Dolichandroue falcata and 
others. In fact any tree or shrub found in our area may occur. 
VII . Other types of Savannah on Sandy Soil. 
The vagaries of the associations found on sand are so numerous that 
different types cannot be distinguished. We regard such associations as 
belonging to true Savannah, with the reservations made above and more 
fully elaborated below. One of the difficulties in the way of a natural 
classification of the sand associations is the human element, the sand area 
being more densely populated and more extensively cultivated than the 
associations just described. 
The following is a fairly complete list of plants occurring in this 
formation : — 
