e on 
— um 
stan- 
Ss 
llec- 
ans- 
as 
r- 
72 h 
was 
for- 
ireen 
-an- 
ted 
lected 
le 
65, 
, the 
1gth 
id, 
n 
iS 
3sed 
ments 
1960) 
sed 
oeri- 
pllec- 
ts on 
nces 
hirsu- 
on- 
ion 
darker- 
rea 
arger 
wa- 
nesses 
tis- 
     
  
  
  
  
   
  
  
  
   
  
  
  
  
  
  
  
Stressed leaves had lower reflectance than nonstressed leaves 
over the entire 0.5- to 2.5-um waveband (Fig. 1A). Within the 
visible spectral region (the 0.55- and 0.65-um wavelengths), 
stressed leaves had lower reflectance than nonstressed leaves 
(P = 0.01), apparently because stressed leaves contained more 
chlorophyll (4.3 mg/g) than nonstressed leaves (4.1 mg/q). At 
the 0.85-um wavelength (within the near-infrared spectral 
region), stressed leaves had significantly (P = 0.01) lower 
reflectance than nonstressed leaves. 
Stressed leaves (Fig. 2B) showed no evidence of abnormal cells, 
but they had compact cellular arrangement in the mesophyll with 
few intercellular spaces, whereas nonstressed leaves (Fig. 2A) 
had a loosely arranged (spongy) mesophyll with many intercellular 
spaces. Thus, the lower reflectance of stressed leaves is associ- 
ated with a compact mesophyll, and the higher reflectance of 
nonstressed leaves is associated with a spongy mesophyll. The 
near-infrared reflectance difference between nonstressed and 
stressed leaves apparently was caused by differences in internal 
cellular structure of the leaf mesophyll. 
Over the 1.35- to 2.5-um waveband, stressed leaves had signifi- 
cantly (P = 0.01) less reflectance than nonstressed leaves, 
apparently because stressed leaves had higher water contents. 
Both stressed and nonstressed plant canopies completely obscured 
the soil within the field of view that measurements were made. 
Stressed plants had a lower reflectance than nonstressed plants 
over the entire 0.5- to 2.5-um waveband. The lower reflectance 
of the stunted, stressed plants (Fig. 1B) as compared with non- 
stressed plants in the visible, near-infrared, and infrared 
water absorption regions was primarily caused by their darker- 
green foliage, smaller leaves with a more compact internal struc- 
ture, and more succulent foliage, respectively. Hence, spectrora- 
diometric field data supported the spectrophotometric laboratory 
reflectance measurements on leaves collected from the field. 
Our data showed that leaves of nematode-stressed cotton plants 
have less reflectance than leaves of nonstressed plants over the 
entire 0.5- to 2.5-pm waveband. The reflected spectral responses 
of leaves from plants grown under different stress conditions 
(like nematode infestation, salinity stress, water stress, nutrient 
deficiencies, insect infestations, and diseases) must be known. 
In remote sensing, an awareness of these reflectance characteris- 
tics should facilitate detecting stressed plants and distinguishing 
them from normal plants. Such studies promote a better understan- 
ding about the reflectance produced by stressed-plant leaves. 
Our preliminary results indicated that remote sensing has much 
potential for distinguishing nematode-infected from noninfected 
plants. 
Iron-deficiency (Gausman et al., 1975b) 
We conducted this study to determine if multispectral data from 
ERTS-1 (now LANDSAT-1) could be used to detect differences between 
chlorotic (iron-deficient) and green (normal) grain sorghum 
(Sorghum biocolor (L.) Moench) plants.