How to Make an Index

by Henry B. Wheatley

"Stickphast" is a very good material; it sticks well and keeps well, and it is an excellent adjunct to the writing-table, but it is not suitable for pasting down a long index. It is too dear, it is too thick, and it is too lumpy. If the paste is made at home, it need not be lumpy; and lumps, when you are pasting, are irritating to the last degree.

The paper and the paste being ready, with a fair-sized brush to spread the paste, we come to consider how best to proceed with the work in hand. You require a good-sized table,—a large board on tressels in an empty room is the best, but a dining-table will serve. At the extreme right of the table you place the batch of paper upon which you are about to paste, and then sort your slips in perfect order, ranging them in columns from right to left. The object of thus going backwards is to save you from passing over several columns as you take the slips off the table, and, instead, going straight on. You can push your batch of paper on as the various columns successively disappear. More slips should not be set out than you can paste at one sitting, as it is not well to leave the slips loose on the table. Of course, you can paste from the left side if you wish, and then the columns will range from left to right; but this is not so convenient for continued arrangement of the columns of slips as you require them.

There are more ways than one in placing the paste upon the paper; the most usual way is to paste down the two sides of the paper just the width of the slips, and some add a stroke down the middle. Another way is to put a plentiful supply of paste on a page or board, and then to place the back of each slip upon this. If you place your fingers on the two ends and press them towards the middle, the slip will be ready to be placed in its proper position, having taken up just sufficient paste. A still different plan is to paste the board or paper as in the previous case, and then place the face of the whole page on this. You then take it off, and, placing the dry side on the batch of paper, proceed to affix the slips to it. The advantage of the two last processes is that the paper is not so wet as in the first-mentioned plan, and in consequence the paper does not curl so much, but lies flatter. In the first place the sheets must be set out separately on the floor to dry, so that they may not stick together, but this is not so necessary in the two latter processes.

Some indexers strongly object to pasting. This was the case with Mr. E. H. Malcolm, who wrote thus to Notes and Queries:

"I long ago discovered the cause of imperfections in my own work. It was the 'cutting into slips' and 'laying down' processes. The fact is you cannot be sure of preserving the cuttings or slips, if very numerous; they are almost certain to get mixed or lost, or elude you somehow. My remedy is this. I now take cheap notepaper and write one entry only on each leaf. Having compiled my index thus from A to Z, I arrange my slips and manipulate them as I would a pack of cards, although shuffling only for the purpose of getting the arrangement of the letters right. Thus I save myself all the labour and trouble of pasting or laying down the slips in analytical order. I do not mind a little extra expenditure of paper by only entering one item on every slip, for I am compensated for the appearance of bulk by finding that I have secured order and arrangement free from the consequences of a finical arrangement of the slips and a dirty and tiresome labour of pasting down." [21]

[ [21] 5th S., vi. 114 (1876).

As already pointed out in these pages, Mr. Malcolm is quite right respecting slips for a growing index; but when it comes to sending the "copy" to the printer the case is different. Here there is more safety in the pasted down slips, which are less likely to be lost than the loose ones even when numbered.

As you proceed in your work you may wish to know how far your index agrees with other indexes in its proportion of letters, and to calculate what proportion of the whole you have already done.

Some calculations as to the relative extent of the different letters have been made. Thus B is the largest letter in an index of proper names, but loses its pre-eminence in an index of subjects; and S takes high rank in both classes.

Mr. F. A. Curtis, [22] of the Eagle Insurance Office, made in 1858 a calculation of the relative proportions of the different letters of the alphabet in respect to proper names. He described his object in a letter entitled, "On the Best Method of Constructing an Index." He wrote that, having had occasion to construct an index of the lives assured in the "Eagle" Company, he had drawn up a few observations upon the subject. "The requirements of an index and the proportions of its several parts are the two principal questions to be considered. Under the first head it may be observed that the index of a company upon a large scale should afford as much abstract information as possible. Those who refer to it do so with different views, for the objects of their inquiry must necessarily vary with their respective duties. It is therefore desirable that the index should be constructed with a view to provide for the wants of each person, so far, at least, as to enable him to obtain information in the most direct way; and it will be proper to insert in the index particulars some of which do not usually find a place in such a book. Let it be supposed that an individual signing his name 'J. Smith' inquires about the bonus, premium, or assignment, etc., of his policy, without stating either number, date, or amount. This is not an unusual case, and it will serve to illustrate my meaning by showing the nature of the difficulties which have to be encountered. J. may stand for John, James, Joseph, etc. There will probably be many of each kind in connection with the like surname, and it would be very difficult to discover, without a tedious investigation, to which policy J. Smith refers, unless the individuality of each person recorded in the index under that name be distinctly shown. The 'locality' of the assurance might be adopted as a mark of distinction; and we should in many instances be able to fix upon the right name by simply comparing the address of the writer with the place where the policy was effected."

[ [22] Assurance Magazine, vol. viii., 1860, pp. 54-7.