,
of such length and arrangement as to form a symmetrical figure or
flower. They do not always take the same form, but follow the same laws
that govern the formation of ice crystals. The structure of a snowflake
may be often found upon a window pane of a frosty morning. Here,
however, the free arrangement of the parts of a snow crystal are
interfered with by its contact with the window pane, but while floating
gently in the air there is the utmost freedom for the play of nature's
forces as they apply to the work of crystallization.

The difference in structure of snowflakes is chiefly due to the
conditions under which they are formed. If the moisture is frozen too
rapidly the molecular forces that are active in crystallization do not
have time to carry out the work, in its completeness of detail, as it
will where the freezing process, as well as the condensing process, goes
on more slowly.

CHAPTER XVI.

METEORS.

Meteors are the tramps of interplanetary space. They sometimes try to
steal a ride on the surface of the earth, but meet with certain
destruction the moment they come within the aerial picket line of our
world's defense against these wandering vagrants of the air. They have
made many attempts to take this earth by storm, as it were, and many
more will be made. They fire their missiles at us by the millions every
year with a speed that is incredible, but thanks to the protecting
influence of the great ocean of air that envelops our globe they become
the victims of their own velocity.

Meteors or shooting stars are as old as the earth itself, and they are
the material of which comets are made. Before it was determined what
these meteors or shooting stars were, many theories were promulgated as
to their origin. One was that they were masses of matter, large and
small, projected by volcanic action from the face of the moon with such
violence as to be brought within the attraction of the earth. Others
supposed them to be the effect of certain phosphoric fluids that
emanated from the earth and took fire in the upper regions of the
atmosphere. This, however, was mere speculation and without any
scientific basis of fact. Anyone who has been an observer of shooting
stars will have learned that there are certain periods of the year when
they are more numerous than at other times; notably in August and
November. Then again there are longer periods of many years apart. By
persistent observation it has been