Losses should not generally be taken on the long side

in a bull period. Nor should they generally be taken on ude the short side in a bear period. It is a bull period as long as the average of one high point exceeds that of previous high points. It is a bear period when the low point becomes lower than the previous low points. It is often difficult to judge whether the end of an advance has come because the movement of prices is that which would occur if the main tendency had changed. Yet, it may only be an unusually pronounced secondary movement.

The first thing for any operator to consider is the value of the stock in which he proposes to trade. The second is to determine the direction of the main movement of prices. We know of nothing more instructive on this point than the course of prices as printed daily.* The third thing is to determine the position of the secondary swing.

Assume for instance that the stock selected was Union Pacific; that the course of prices afforded clear evidence of a bull market under way; that the high point in Union Pacific thirty days ago was 108; that the price had slowly declined in sympathy with the market and without special new features to 98. The chances would be in favor of buying a part of the line wanted at that price with the intention of buying a little more if the stock had further decline or if the price showed a well defined advancing tendency. It would then be wise to watch the general market and wait for an advance.

A 10-point decline under such conditions would be almost certain to bring in a bull market more than 5 points *See Wall Street Journal.

recovery and full 10 points would not be unreasonable; hence if the general market maintained a good tone, it would be wise to wait for 5 points and then begin to think about stop orders.

Even in a bear market, this method of trading will usually be found safe, although the profit taken should be less because of the liability of weak spots breaking out and checking the general rise.

CHAPTER IX.

*METHODS OF READING THE MARKET.

A correspondent writes: "Is there any way of forecasting the course of the market from the tape, from your records of transactions or from the summarized movement of prices? Transactions must mean something, but how can a trader tell what they mean?"

This is an old question. There have been a variety of answers but it is doubtful if any have been or can be wholly satisfactory. Several methods, however, are in practical use and at times afford suggestions.

There is what is called the book method. Prices are set down, giving each change of 1 point as it occurs, forming thereby lines having a general horizontal direction but running into diagonals as the market moves up and down. There come times when a stock with a good degree of activity will stay within a narrow range of prices, say 2 points, until there has formed quite a long horizontal line of these figures. The formation of such a line sometimes suggests that stock has been accumulated or distributed, and this leads other people to buy or sell at the same time. Records of this kind kept for the last fifteen years seem to support the theory that the manipulation necessary to acquire stock is often times detected in this way.

Another method is what is called the theory of double

*Dow's Theory.

tops. Records of trading show that in many cases when a stock reaches top it will have a moderate decline and then go back again to near the highest figures. If after such a move, the price again recedes, it is liable to decline some distance.

Those, however, who attempt to trade on this theory alone find a good many exceptions and a good many times when signals are not given.

There are those who trade on the theory of averages. It is true that in a considerable period of time the market has about as many days of advance as it has of decline. If there come a series of days of advance, there will almost surely come the balancing days of decline.

The trouble with this system is that the small swings are always part of the larger swings, and while the tendency of events equally liable to happen is always toward equality, it is also true that every combination possible is liable to occur, and there frequently come long swings, or, in the case of stock trading, an extraordinary number of days of advance or decline which fit properly into the theory when regarded on a long scale, but which are calculated to upset any operations based on the expectation of a series of short swings.

A much more practicable theory is that founded on the law of action and reaction. It seems to be a fact that a primary movement in the market will generally have a secondary movement in the opposite direction of at least three-eighths of the primary movement. If a stock advances 10 points, it is very likely to have a relapse of 4 points or more. The law seems to hold good no matter

how far the advance goes. A rise of 20 points will not infrequently bring a decline of 8 points or more.

It is impossible to tell in advance the length of any primary movement, but the further it goes, the greater the reaction when it comes, hence the more certainty of being able to trade successfully on that reaction.

A method employed by some operators of large experience is that of responses. The theory involved is this: The market is always under more or less manipulation. A large operator who is seeking to advance the market does not buy everything on the list, but puts up two or three leading stocks either by legitimate buying or by manipulation. He then watches the effect on the other stocks. If sentiment is bullish, and people are disposed to take hold, those who see this rise in two or three stocks immediately begin to buy other stocks and the market rises to a higher level. This is the public response, and is an indication that the leading stocks will be given another lift and that the general market will follow.

If, however, leading stocks are advanced and others do not follow, it is evidence that the public is not disposed to buy. As soon as this is clear the attempt to advance prices is generally discontinued. This method is employed more particularly by those who watch the tape. But it can be read at the close of the day in our record of transactions by seeing what stocks were put up within specified hours and whether the general market followed or not. The best way of reading the market is to read from the standpoint of values. The market is not like a balloon plunging hither and thither in the wind. As a whole, it represents