better organized, "tradesunionism tends to increase the size of the strike." We can consider this proposition from two different standpoints, (1) from that of unions in the aggregate and (2) from that of unionism. To consider the case of unions in the aggregate, we shall have recourse to statistics such as Huebner uses, but in the case of unionism mere analysis must suffice, as there are no available statistics.

DIAGRAM VI. CHANGES IN THE AVERAGE NUMBER OF ESTABLISH

We have already seen that the average strike has not become more wide-spread. Diagrams VI and VII indicate what has happened in regard to union and non-union strikes, considered in the aggregate. Diagram VI shows that the average number of establishments affected per non-union strike steadily decreased from 1881 to 1905. The curve representing the average number of establishments per union strike shows a marked decrease from 1881 to 1891, a period of active organization and great strength on the part of the Knights of Labor. During the depression of 1893 and the consequent inactivity on the part of the unions the number of establishments per union strike increased very rapidly. From 1897 to 1905, however, during which time tradesunionism grew by leaps and bounds and became more experienced, the average decreased rapidly. Similar results are to be noted in the case of the average number of strikers affected per union and per non-union strike, as shown by Diagram VII. The level reached in this case is lower in 1904 than in any other year during the period under discussion. Thus we conclude that the average union strike does not tend to increase in size, nor does it become more wide-spread with the passing of years and the growth of experience on the part of unions in the aggregate.* It is difficult to see why the situation should be otherwise. In the first place, as long as strikes remain more or less local in nature, they will continue to be rather small in size. It is only when the cause of the trouble between the employer and the employees is wide-spread, as in the case of an eight-hour agitation, railway strikes, etc., that a very large number of men are affected. ondly, the working class is organized to-day for the most part upon trade lines, there being comparatively few industrial unions. With each trade in a shop organized into a separate union, it is easy to see that the number of men in that factory belonging to each union is, as a rule, small. Thus, whenever a strike is called and the union quits work, a very few men only

Sec

* More accurate results could be obtained, were strikes classified in accordance with the number of strikers, establishments, and employees affected. This is partly done in Germany, where all strikes are classified as involving 2 to 5, 6 to 10, 11 to 20, 21 to 30, 31 to 50, 51 to 100, 101 to 200, 201 to 500, and 501 or more individuals.

are affected. The rest of the laborers in the city who belong to that union do not strike, neither do the men in other trades in the same factory. It is only the men who are employed at that particular trade in that particular shop who go out

DIAGRAM VII. CHANGES IN THE AVERAGE NUMBER OF STRIKERS

upon the strike.* Thirdly, with the growing complexity of industry the old trades are being broken up into minute subdivisions, each with its separate union. In times past all of these subdivisions belonged to the same trade and considered that they had common interests. Consequently, they struck together. But at the present time, if one of these subdivisions has a grievance, it strikes of its own accord, thus making the average union strike, of necessity, a small one. Fourthly, the change in the causes of industrial disputes is without doubt the most important reason for the decrease in the size of the union strike. During the last ten years, as we have already seen (Diagram I), the closed-shop policy and union rules have rapidly increased as causes of strikes, while wages and hours have been decreasing in importance. A strike ordered because of a demand for a closed shop or for the recognition of the union seldom affects more than one establishment because the grievance is purely local in its nature, and, inasmuch as these strikes have greatly increased in number, they have tended to diminish the size of the average union strike. The closed shop is a demand arising from old and experienced unions. It is not a grievance for the redress of which unions are formed, nor is it one that causes newly organized trades to strike. Its increasing importance as a cause of union strikes is in itself a strong refutation of Huebner's contention that unionism tends to greatly increase the size of the strike.

Diagram VIII gives the percentage of establishments in which strikes were (1) entirely successful and (2) partly successful. From the facts presented we learn that strikes were less successful in 1902, 1903, and 1904 than ever before, and that the percentage of compromised or partly successful strikes increased slowly during the period under discussion.

Diagram IX gives the percentage of establishments in which strikes failed and in which union and non-union strikes failed. A steady increase is to be noted after 1890 in the number of

*The growth in the number of sympathetic strikes is, however, of significance in this connection.

non-union strikes that were unsuccessful. The percentage of union strikes that failed increased steadily during periods of business activity and decreased during business depressions. The tendency of union strikes to become less successful since 1898 substantiates Huebner's statement that "union strikes are not becoming more successful even though unionism is being