METEORS AS PROBES OF THE UPPER ATMOSPHERE 
observed meteoroids are probably of a different struc- 
ture than meteorites. 
The studies of meteor spectra, largely by Millman 
[55, 56, 57], indicate predominance of low-excitation 
Fet lines, with Fer lines [85] possibly showing in one 
case, probably for a very high-velocity meteor. Present 
also are lines of the atoms Nat, Cat, Mg1, Mnt, Crt, 
Si, Ni, Alt, Cau, Mgt, Sim, and possibly Feo. No 
atmospheric constituents have been observed. The rela- 
tively low states of excitation correspond to tempera- 
tures of only a few thousand degrees, in spite of the high 
energies of the impinging atoms; for example, a nitro- 
gen atom at a velocity of 70 km sec carries an energy 
of 410 ev. 
The average meteoroid is probably an irregularly 
shaped stone or stony iron. There is no reason to believe 
that it is structurally strong or homogeneous in physi- 
cal or chemical structure. Roughly 3 per cent of the 
meteors photographed by Harvard split into two or 
more pieces in the upper atmosphere, while nearly 10 
per cent showed flares in brightness. Jacchia [35] has 
shown definitely that flares arise from quick losses of 
material, probably by crumbling or breakage of the 
meteoroid. Direct evidence as to the nature of meteor- 
oids may sometime be obtained from micrometeorites 
[38, 92], those meteoroids that are sufficiently small to 
be stopped by the atmosphere without vaporizing. 
Until then we can only postulate the structure of 
meteoroids and check our postulates indirectly by var- 
ious observations and deductions. An acceptable theory 
as to the nature and origin of comets would also be of 
value in the study of meteors. 
ATMOSPHERIC RESEARCH BASED ON 
VISUAL OBSERVATIONS 
Visual Observations of Meteors. As early as 1798 
Brandis and Benzenberg set out to determine the 
heights of meteors by simultaneous observations from 
two separated stations. In all they observed 402 
meteors, of which 22 appeared to be identical. Newton 
[60], a major contributor to early meteor studies, ana- 
lyzed 21 of these ‘‘pairs” and calculated heights, rang- 
ing from 12 to 245 km, with a mean of about 100 km. 
The mean is in good agreement with modern measures 
but the range is far too great. Average visual meteors 
become observable at a height of about 100-110 km 
with a large scatter, and persist to 90 km or lower, de- 
pending upon their brightness. High-speed photo- 
graphic meteors first show at greater heights, about 120 
km, while the chief activity observed by electronic 
techniques is generally near the H-layer. The slowest 
and brightest photographic meteors observed at Har- 
vard disappear in the neighborhood of 40 km. 
Lindemann and Dobson [41, 43] made the pioneer 
application of meteoric theory and observation to a 
study of the density and temperature of the upper 
atmosphere. They developed the first comprehensive 
theory of the meteoric process and then utilized the ex- 
tensive visual observations of meteor heights that had 
been made by that most assiduous of meteor observers, 
W. F. Denning, and by his co-workers. Since Mitra 
307 
[59] has recently reproduced in extenso the theoretical 
developments by Lindemann and Dobson, only a few 
remarks concerning their work will be made in this 
article. Briefly, Lindemann and Dobson recognized and 
stated clearly the fundamental meteoric processes, 
namely, surface heating by impact with air molecules, 
stressing the importance of effective mean free path, 
vaporization of the meteoroid, luminosity produced by 
encounter of the vaporized material with the air, and 
the relatively slow deceleration of the nucleus remain- 
ing. 
Although their thermodynamic, rather than kinetic, 
approach to the problem of heat transfer through the 
gas cap led to erroneously high values of the calculated 
air densities at great altitudes, as pointed out by 
Sparrow [81] and later workers, nevertheless the basic 
processes as visualized by Lindemann and Dobson are 
in other respects fundamental. Progress towards a com- 
pletely satisfactory theory is still surprisingly slow. 
Lindemann and Dobson showed clearly that the 
previous concept of a constant stratospheric tempera- 
ture in the upper atmosphere must be replaced by the 
recognition of higher temperatures at great altitudes. 
They noted further that from Denning’s observations 
the end heights of meteors were markedly less frequent 
in the region from 50 to 60 km than immediately above 
or below. From this fact they concluded that the at- 
mospheric temperature must rise abruptly at a height 
of about 60 km so that ‘‘As the meteor passes into colder 
air the temperature of the air cap will fall so that the 
heating will be reduced.” This latter argument and 
consequent conclusion must be revised slightly. It is 
more accurate to conclude that a region of maximum 
temperature must be present between 50 and 60 km 
because the existence of a smaller logarithmic density 
eradient prolongs the lifetime of a meteoroid reaching 
this height; the meteoroid has, therefore, a smaller 
chance of disappearing in the critical range of altitude. 
The air temperature, per se, is not the dominating fac- 
tor; it is the consequent reduced density gradient which 
is important. 
Related effects of a variable logarithmic gradient im 
atmospheric density show markedly in meteor trails 
photographed at Harvard and account largely for the 
phenomenon observed by Hoffleit [29] (see also [23}) 
that the point of maximum brightness moves systemati- 
cally forward in meteor trails with increasing velocity 
(for discussion see [87]). 
Lindemann and Dobson also pointed out the fact 
that meteor heights are systematically greater during 
the summer than during the winter. They could not 
conclude positively, however, that the height of a poimt 
of given density in the atmosphere is greater in summer 
than in winter. It is possible that there are systematic 
differences in velocity between the seasons and that 
the heights of meteors are very dependent upon velocity. 
These general problems concerning visually observed 
meteors still constitute a source of considerable dis- 
cussion or disagreement [44, 51, 72, 73]. 
The Arizona Expedition for the Study of Meteors 
was planned by Shapley, Opik, and Boothroyd [77]. 
