cavities corroded and formed by 
chemical reaction at the expense of 
the gehlenite. In order to study the 
material of the cavities more closely, 
photographs were taken with a scan- 
ning electron microscope and the min- 
erals of that material were identified 
from the energies of the X-rays emit- 
ted by them under electron bombard- 
ment. The minerals in the veins and 
cavities form beautiful, well-devel- 
oped, multifaced crystals with large 
amounts of empty space between 
them. Textures with such empty 
spaces are usually interpreted as im- 
plying crystallization in a gas pocket, 
a process in which the solid products 
occupy only a small fraction of the 
volume of the starting vapor. Although 
the minerals of these veins and cavities 
formed by reaction of a previously 
condensed gehlenite with a gas, none 
of them are predicted by thermody- 
namic calculations to condense from 
a gas of solar composition. One pos- 
sibility is that the solar nebula differed 
in chemical composition from place 
to place and that the Allende inclu- 
sions condensed in one region of solar 
composition but were altered in an- 
other area of different composition. 
Another indication that the history 
of the coarse-grained inclusions did 
not end with the condensation of the 
major minerals in their interiors is the 
presence of rim layers around the out- 
sides of these inclusions. Every inclu- 
sion is surrounded by the same se- 
quence of four mineralogically distinct 
rims, each of which is so narrow (three 
to thirty microns) that a scanning elec- 
tron microscope is required for its ade- 
quate study. The innermost zone is 
made of typical high-temperature con- 
densates, but the next layer is rich 
in one of the alteration minerals seen 
in the veins and that was formed by 
reaction between the previously con- 
densed minerals and the surrounding 
nebula. The third layer from the in- 
terior of the inclusion consists of an- 
other alteration mineral that grades 
outward in composition to another 
mineral formed by later condensation. 
The outermost layer is made of two 
minerals rich in iron oxide. 
The rims have several noteworthy 
features. Because they are not granu- 
lar, they cannot have formed by the 
accretion of grains that crystallized 
elsewhere, as in sedimentary rocks. 
Rather, their structure suggests that 
they crystallized in situ, perhaps by 
reaction between inclusion interiors 
and some other material. This reacting 
material cannot have been minerals 
in the meteorite matrix in which the 
inclusions are embedded. The reason 
is that such a reaction would produce 
a rim all the way around each in- 
clusion, but sometimes the rims ex- 
tend only part way around an inclusion 
even though the matrix completely 
surrounds the inclusion. Rimming 
must have occurred prior to the in- 
corporation of the inclusions into the 
meteorite’s parent body, but some 
rims were broken off between these 
two events. The rims probably formed 
in a reaction between the inclusions 
and the gas of the solar nebula, but 
some of the reaction products cannot 
form at chemical equilibrium in a gas 
of solar composition. Thus, there is 
Type B inclusions, as this example 
photographed with a light 
microscope shows, are nearly 
spherical. Their shape and tightly 
interlocking grains suggest that they 
were once molten droplets. The 
mineral around the edges here is 
gehlenite. Other smaller inclusions 
are visible in the surrounding 
matrix. 
R S. Kjarval 
once again the possibility that the in- 
clusions were formed in a region of 
nonsolar composition. The presence of 
iron oxides in some rim minerals in- 
dicates that this region was much 
more oxidizing than the one from 
which the interiors of the inclusions 
condensed because, in it, we know me- 
tallic iron was the stable iron-bearing 
mineral. 
In conclusion, we have seen that 
the Allende meteorite contains objects 
that crystallized in the cosmic cloud 
from which the solar system formed. 
Some are aggregates of mineral grains 
that condensed directly from the gas 
of that cloud. These inclusions are our 
best clues to conditions and processes 
that occurred in that distant era before 
the planets formed. Recorded in the 
inclusions is a rather complex history 
of successive condensation events and 
evidence that the cloud was hetero- 
geneous in chemical composition. 
Data show that the cloud was also 
isotopically heterogeneous and that 
collapse of the interstellar cloud to 
form the solar system may have been 
triggered by the explosion of a nearby 
supernova. It seems likely that the 
inferred spatial heterogeneity of 
chemical and isotopic composition of 
the nebula was caused by last-minute 
injection into the nebula of supernova 
matter of different composition. Dif- 
ferent parts of individual inclusions 
appear to have reacted with gases of 
different chemical and isotopic com- 
position, apparently reflecting the fact 
that the inclusions began forming be- 
fore newly injected material was thor- 
oughly stirred with the older matter. □ 
71 
