167 
Table 10 
Summary of botanical composition of 20 esophageal 
fistula collections from reindeer at Prudhoe Bay. 
(a) Percentage occurrence of plant groups and _ parts 
present in more than 1% of fistula samples. 
Cyperaceae 54.6 
green (37.9) 
dead (16.7) 
Salicaceae 28.7 
green (27.1) 
dead ( 1.6) 
Gramineae green 1.9 
Herbs 733) 
Lichens 2p) 
Vascular bundles unidentified 45 
Other dead unidentified 3.6 
100.0 
Cyperaceae constituted more than 50% of the 
diet, while Salicaceae composed almost 30%. 
Herbs formed only 7.5% of the diet, while the 
contribution of lichen and Gramineae was in- 
significant. Table 10(b) shows the frequency of 
occurrence of the main plant species. Analysis of 
esophageal egesta indicates that FE. angustifolium 
was the most frequently eaten plant during the 
summer season. 
(c) Botanical composition of caribou rumen 
samples. During the last week of July, four 
caribou rumen samples were collected for botan- 
ical analysis (Table 11). Almost 35% of the 
sample was Salicaceae; 28% Cyperaceae (identifi- 
able were 8% Eriophorum spp., 3% Carex 
aquatilis), and the balance was composed of 
herbs, lichen, and dead material. Approximately 
14% of the sample was dead material. The 
nutritional history of the caribou was uncertain, 
but they were observed to graze mainly the 
Dupontia brook-meadow and, possibly, the sur- 
rounding Eriophorum marsh types preceding 
capture. From this standpoint, the analyses may 
be compared with esophageal egesta analyses 
obtained from reindeer grazing a Dupontia 
brook community at the same time. Comparison 
of values for the rumen samples in Appendix 
Table 5 with the values from the EF samples in 
Fig. 7 (Dupontia meadow) shows a_ greater 
amount of grass-like groups compared to willows 
in the EF samples, while the herb group is about 
equal. The caribou rumen samples show a higher 
(b) Percentage occurrence of plant species present in 
more than 1% of fistula samples. 
Eriophorum angustifolium 36.6 
Salix ovalifolia 9.0 
Carex scirpoidea 6.6 
Salix arctica 5}5// 
Salix rotundifolia 3E5 
Carex aquatilis $15) 
Dupontia fisheri 1.8 
Dryas integrifolia 1.8 
Thamnolia vermicularis 1.8 
Carex rupestris lESeevaleo 
Trace and unidentified species 28.4 
100.0 
content of lichen than the EF samples. Detec- 
tion of quantitative differences in composition 
between the rumen and EF samples cannot be 
made due to the small sample sizes. Also, these 
differences in composition could be related to 
differential rates of ruminal turnover and/or 
digestion of the plant species. An absolute dif- 
ference in selective patterns of reindeer and 
caribou cannot be determined, although prefer- 
ence for Sa/ix spp. was apparently made by 
caribou. 
Plant biomass estimates in relation to vegetation 
type and grazing behavior 
Table 12 lists estimates of live, dead, and 
total biomass on five study sites at Prudhoe Bay. 
Peak biomass was recorded for each vegetation 
type in early to mid-July 1973. A marked 
decline in live biomass was noted in August. 
Insufficient sampling early in the season prevent- 
ed the construction of precise growth curves for 
these sites. However, based on previous studies 
of the seasonal progression of primary produc- 
tion at Barrow, Alaska, a linear growth is expect- 
ed from 20 June. Predicted seasonal patterns in 
primary production for the main vegetation 
types at Prudhoe Bay are shown in Fig. 9. 
Unfortunately, we were unable to estimate be- 
tween site variations in the seasonal primary 
production of any one vegetation type. Subjec- 
tive estimation suggests that variation within 
most vegetation types was high. For example, 
