April 25. What was measured has grown deeper.
At half past nine I looked at the screen after the combination was confirmed.
It was confirmed on the twentieth of April. The set stands in the list as one row.
Open that row and the five sub-numbers are inside it. A man who opens one sees five.
Twenty days ago I wrote the application wanting it to be so. It is so.
The imprint box says unknown. The compiler box is not in the list.
The title can be seen.
The name Pnakotic Press has come up on the outside screen. It does not catch in the index. It can be seen as a title.
At ten to ten I ran off the query record.
Over five days the set's row has been opened twenty-nine times.
I looked at the account boxes. Twenty-six say outside. Three carry eight Latin characters.
They are the eight that came up once on the fourth of March. The side that sent the enquiry.
The set has stood five days and that person has opened it three times.
From ten I checked sub-number two.
The surveyor's field book. Two hundred and six pages, the thickest of the five.
Each line has a station number and a sounding and the date it was taken. There are two thousand three hundred and forty-two soundings.
I copied it by hand over eleven days.
From half past ten to twelve I matched the figures one by one.
For eleven days I copied numbers.
What I copied three days ago was words. Words have meaning, so an error catches the eye. One character wrong and it does not make sense.
Numbers do not catch. Three and eight exchanged, a number is still a number. Reading it there is no way to know it is wrong.
So I wrote each twice and went along matching the two. That is why it took eleven days.
The two writings agree with each other. They do not agree with the original.
The count is the same. Two thousand three hundred and forty-two, and the copy two thousand three hundred and forty-two.
Twenty-three values differ.
All twenty-three are larger in the copy.
Not one is smaller.
From half past twelve to one I looked at what the larger values are.
I set the twenty-three against the original one at a time.
All twenty-three equal the value at the same position on the original's following page.
A number that is one page later has come forward.
At ten past one I looked at the last page.
One of the twenty-three is on page two hundred and six.
Page two hundred and six has no following page. The field book ends there.
In the copy there is a value in that position.
That value is nowhere in the original.
I went through all two thousand three hundred and forty-two. It is not there.
I looked at the value again.
It is larger than the largest of the original's two thousand three hundred and forty-two.
I found the deepest place in the original and set it against it. It is deeper than that.
The deepest number in this field book is a number that was not in this field book.
That position is the last line of the last page.
Why the field book ends there is not written. Under the last page those ten characters are stamped, and directly above them is that line.
The ten characters that are on the last page of all four.
Directly above those ten characters is a number I wrote.
From half past one to half past two I looked up what I wrote eight months ago.
There is a rule this room first wrote in August.
What diverges is always the copy.
The copy then was made by the machine. The machine wrote a day early and the machine's number was larger.
Today's copy, on the eleventh day, was made by my hand.
Three days ago the librarian's was my hand too.
Twice is not once.
From half past two to half past three I looked at what growing larger means.
A sounding is a depth. A larger figure is deeper water.
At twenty-three stations the copy's water is deeper than the original's.
In the record of the man who measured, the water has grown deeper. The man who measured stood there sixty-six years ago and measured that depth then.
There is no way in this room to know which would come out if one went and measured now.
Nor can one go. The field book has station numbers only, and where they are is in another register. That register is not in this room.
At twenty to four I looked at where the twenty-three are.
They are spread evenly over two hundred and six pages. Eight near the front, seven in the middle, eight at the back.
I divided them by which of the eleven days I wrote them on. That is even too.
They are not bunched where I tired. There are some on the first day and some on the last.
At ten to four I counted the dates of the twenty-three.
Eleven are December.
I counted the whole. Of the two thousand three hundred and forty-two, a hundred and ninety-four are December.
About eight in a hundred.
Eleven of twenty-three is forty-eight in a hundred.
By counting it is many.
I did not write what its being many means. There is no box for it, and writing it leans one way.
At ten past four I opened the right-hand material.
Three hundred and thirty-three pages. The date of the added page is 24 May 98.
My log is on its three hundred and thirty-third page today.
At half past four I opened the box.
Seven. Six impressions. The sixty-first day.
In July it took a hundred and seven days for the impressions to become six and the items to become six. Today is the sixty-first, so by that count forty-six remain.
I know that waiting by a count is not counting. I am counting all the same.
At five I looked at the sixth box.
Six hundred and thirty-eight.
At ten past five I wrote the ten-line summary for sub-number two.
Two hundred and six pages, two thousand three hundred and forty-two soundings, I wrote.
The count being the same, that line does not differ between original and copy.
That twenty-three values differ does not go into ten lines.
The librarian's split in count. The surveyor's grew in value.
Both on the copy's side. The place where they increase is different.
One gained a row. In the other the rows stayed and the numbers inside them grew.
Next is the lighthouse keeper's table of tides.
A tide is not a count and not a value. It is an order.
How an order increases I do not know.