deciphered it. I have a hope that I may gain some light from your
discovery, and thus be able to do something myself."

"Well," said Miss Krieff, "I will tell you, since you have failed so
completely. My principle is a simple one; and my deciphering, though
only partial, seems to me to be so true, as far as it goes, that I
can not imagine how any other result can be found.

"I am aware," she continued, "that there are forty different
characters in the inscription. I counted them all out, and wrote them
out most carefully. I went on the simple principle that the writer
had written in English, and that the number of the letters might be
disregarded on a first examination.

"Then I examined the number of times in which each letter occurred. I
found that the sign [Aries image] occurred most frequently. Next was
[Gemini image]; next [Taurus image]; and then [Cancer image], and
[Leo image], and [Libra image], and [Sagittarius image], and [Mars
image]." Miss Krieff marked these signs down as she spoke.

Gualtier nodded.

"There was this peculiarity about these signs," said Miss Krieff,
"that they occurred all through the writing, while the others
occurred some in the first half and some in the second. For this
inscription is very peculiar in this respect. It is only in the
second half that the signs of punctuation occur. The signs of the
first half are all astronomical.

"You must remember," continued Miss Krieff, "that I did not think of
any other language than the English. The idea of its being any
dialect of the Hindustani never entered my head. So I went on this
foundation, and naturally the first thought that came to me was, what
letters are there in English which occur most frequently? It seemed
to me if I could find this out I might obtain some key, partially, at
any rate, to the letters which occurred so frequently in this
writing.

"I had plenty of time and unlimited patience. I took a large number
of different books, written by standard authors, and counted the
letters on several pages of each as they occurred. I think I counted
more than two hundred pages in this way. I began with the vowels, and
counted up the number of times each one occurred. Then I counted the
consonants."

"That never occurred to me," said Gualtier. "Why did you not tell
me?"

"Because I wanted you to decipher it yourself on your own principle.
Of what use would it be if you only followed over my track? You would
then have come on