In this and subsequent sections
we present the numerical results obtained by
the method described in the previous section.
We computed line profiles for
disk sizes of 
5, 10, and 20,
density-gradient indices of 
1, 2, 3, and 4,
characteristic optical depths of 
, 
, and 10
and an optically thin case,
and inclination angles of i = 
, 
, and 
.
Note that observed H
emission arises, on the average,
in the 
range (Slettebak et al. 1992).
Here, we comment on the range of 
estimated for
the disks of actual Be stars.
For shell (
) stars,
the optical depths of the envelopes are
determined by the central depth analysis of
the observed shell absorption lines (e.g., Kogure et al. 1978).
Kogure (1990) collected the results obtained
for some shell stars,
and concluded that the optical depths in H
lines 
range from 2000 to 5000 for fully-developed shell stars,
while 
for weak shell stars.
On the other hand, we can estimate 
for 
in our model as
Hence, we found that
disks of the actual Be stars are characterized
by 
.
Figure 3 shows the various line profiles from the unperturbed disks
for 
.
The left panels are for 
and
the right panels for 
.
The panels are for 
, 
, and 10
from top to bottom, respectively.
In each panel we show profiles for four outer disk radii:
2, 5, 10, and 20.
Each profile is normalized by the peak intensity
of the unperturbed profile from the disk with 
,
, and 
.
Except for the line widths,
the profiles for 
and 
are
roughly the same as those shown in figure 3.
We now consider the effects of the density gradient index 
and the optical depth 
on the line profiles.
As demonstrated by Horne and Marsh (1986),
the intensities along the line-of-sights
through the high-density region saturate in optically-thick cases.
Hence, the area under the saturated line intensity
increases with increasing characteristic optical depth 
.
Simultaneously, the unsaturated intensities from an outer region
of the disk increases with 
.
As a result, the contribution of the outer region of the disk
monotonically increases with 
.
On the other hand, the velocities of the violet and red
peaks for 
are independent of 
, and
are given by 
,
while those for 
decrease towards
as 
increases.
This feature can be understood as follows:
The emission in each velocity bin arises from a different
crescent-shaped region of the disk surface
[see figure 2 and also figure 1 of Horne and Marsh (1986)].
For 
, the strongest emission arises,
irrespective of 
, from the crescent-shaped region
tangent to the outer radius of the disk.
For 
, however,
it arises from the crescent-shaped region
tangent to the radius where the optical depths
along the line-of-sights are almost unity.
This radius increases with increasing 
.
Thus, for 
,
the peak velocity is a measure for a radius
within which most of the emission arises.
