181 
to tree 
nercially 
s rox bur- 
rue e 
combina- 
Jiey occur 
:omous 
les it 
LOW!! 
le charac- 
photographs, but if photographed in other seasons its recognition becomes 
only a guess. On 1:10,000 scale photographs taken in November- 
December of Allapalli forests of Maharashtra, the individual trees of 
Teak in mixture of hardwoods are easily discernible. Of the other 
broadleaved species, the recognition of Salai (Boswellia serrata) , Kulu 
(Sterculia urens), Semal (Salmalia malabarica) have been consistent 
because of the typical tonal contrast and form (size and shape) of the 
trees. E. Van Es and Joshi ( 1 ) claim to have recognised some 
species during the course of inventory of mixed deciduous forests. 
While a set of photo characteristics for recognition of individual species 
is only true for that locality and same type of photographs it has been 
experienced that after a field work of about a fortnight the image of 
some of the species becomes so characteristic that its recognition 
becomes easy. It is more due to the typical form of the species in the 
md 
locality and the coincidence of the season of photography better suited 
to that species. Examples of species like Hardwickia binata, Madhuca 
latifolia, Lagerstroemia parviflora, Diospyrus melanoxylon, Adina cordi- 
2 d crowns 
il differ 
ier and 
are 
folia can be cited. It is however difficult to recognise them off hand 
on the aerial photographs. 
Special mention is made for Bamboos (Dendrocalamus strictus) . 
3 ar very 
P, it 
nat like 
The species economically of great importance to India occurs mostly as 
understory while pure patches in moist deciduous types of forests can 
be met with. If photographed in deciduous season (March-April) appears 
in white tone, star shaped in distinct typical pattern below the leafless 
trees. In open patches in other seasons it can be recognised as light 
grey toned, irregular and wooly texture. Joshi ( 4 ) while working for 
optimum scale of photography for dry deciduous forests of Bastar did not 
and 
find any significant difference in tree species identification between 
photo scales of 1:5,000 to 1:20,000. 
shape. 
ce trees 
ver, the 
because 
"Forest Type Classification" 
Breaking up of a complex forest area into more homogeneous 
units is a prime practice in forest inventories. What should form a 
homogeneous unit depends on the purpose of the survey and the nature 
of the forest. In the initial stages of execution of forest inventories 
of Hard- 
and 
.y on the 
es, 
lleys 
lar 
m o 
good quality aerial photographs on larger scale were not available and 
hence the scale and quality of photos too became one of the factors 
which influenced the limit down to which the forests could be classi 
fied. In conventional ground forestry, forest types are mainly recogni 
sed by CHAMPION and SETH's classification which is mainly of Ecolo 
gical and Silvicultural significance. Adoption of this classification for 
resource inventories was not justified. The development of a standard 
resource based classification for the country is still in process. 
sw spe- 
Different regional inventories adopted one form or the other of the 
classification system depending on the nature of forest, purpose and 
rcially 
jraphed 
well 
ed in 
>f partial 
ppears 
n single 
limitations imposed by photographic prints. However the general scheme 
was to frame a LAND USE classification in the first stage so that forest 
area could be separated for more intensive classification. The broad 
classes were cultivated land including built up area, water bodies, 
forest land etc. Forest area was further stratified into different classes 
of species, height, crown closure, crown diameter, volume. The class 
limits varied in different regions depending on the value of forests and