therefore becomes paramount. Without these statistics, mere cost comparison is a one-legged affair, hard to keep up and easy to get away from.

RESULTS

The writer keenly realizes that he has consumed too much time in discussing some of the problems which make life interesting in this division of the Census. He will therefore take a short cut home in speaking of the concrete results already attained in this work. An imposing story could be told of the beneficient influence of the Census work upon the cause of accounting in Ohio, Indiana, Massacuhsetts, and other states; of the achievements of the two conferences on uniform city accounting held in Washington in 1903 and 1906 at the instance of the Census Bureau, and under the leadership of Dr. Powers; of the numerous addresses and papers which have been presented by officials of the Census Bureau in creating sentiment and enlarging popular interest in the subject of improved accounting; of the many valuable records prepared for cities by field agents; of the annual missionary work done by some of these living pillars of the cause, and, as in other fields of human endeavor, of the backsliders, and the obdurate. This would be a long, but on the whole an encouraging story.

Indirectly also this work has left its mark, though in forms not so easily measurable. The bulletins, though they may fairly claim distant kinship with historical literature, are storehouses of information. Governmental policy demands that they be free from specific critical comment. As a result they present a far more modest and harmless appearance than there contents might provide. And their full utility is apt not to be discovered with out a considerable interpretative study of the data presented.

Summarizing briefly the most vital points aimed at, it would seem that such interest in this subject as you may have, may with profit be directed towards the advancement of the following policies: first, close and harmonious coöperation between city, state and national statistical bodies dealing with municipal statistics, and private organizations interested in civic problems, looking towards the adoption of uniform methods of statistical inquiry, within the limits set by state laws and special conditions; second, the closing of the city accounts and records on one and the same day everywhere, and irrespective of the closing of the fiscal year; third, improvement not only of the recording of financial, but of physical statistics. The

progress and vital results of the work of the Census Bureau will depend largely upon the speed and success with which the significance of these factors is generally recognized, and the effectiveness with which the ideas which they represent are translated into practical achievement. To this task your coöperative interest is invited.

THE DIRECT PRIMARY IN ILLINOIS

BY WALTER CLYDE JONES

Member of State Senate of Illinois

Illinois has had an arduous task in its efforts to enact a valid direct primary law. Unusual conditions exist in Illinois which have made the problem peculiarly difficult.

SPECIAL CONDITIONS IN ILLINOIS

In the first place, Illinois contains at one corner a large city having more than one-third the total population of the state. The political conditions and methods of party organization in Chicago are wholly different from those of the smaller cities and communities of the state. What is suitable for the country districts in the form of a primary law is in many respects unsuited to the metropolitan conditions of Chicago, and vice versa. The country districts, to a great extent, were at the beginning of the movement for direct primaries favorable to a primary which was direct, as distinguished from a delegate-convention system, and were favorable to such a primary of the plurality type, in which a plurality of the votes, not necessarily a majority, should nominate. In Chicago, those active in politics and familiar with party organization were opposed to direct primaries, and particularly antagonistic to a direct primary of the plurality type. They favored a delegate-convention primary system, and in any event believed that a direct primary, if inevitable, should provide for majority nominations.

There was thus a distinct cleavage between the city of Chicago and what we may term the country districts, and this attitude was reflected not only in the party leaders of the respective communities but in the members of the legislature as well.

Another special condition giving rise to difficulties was the principle of minority representation which prevails in Illinois and which is found in no other state in the Union. This system of minority

representation provides for a cumulative vote for members of the House of Representatives of the General Assembly, and this principle of minority representation, with its cumulative voting plan, has presented problems in connection with the drafting of a direct primary law which, in view of the attitude of our Supreme Court, have seemed almost insuperable.

Immediately following the Civil War the northern half of Illinois which had been abolitionist in sentiment, was solidly Republican and the southern half, which had been pro-slavery in sentiment, was solidly Democratic. In consequence, the Democrats of the northern half of the state and the Republicans of the southern half of the state were without representation in the legislature. It was proposed to remedy this condition by the principle of minority representation, which was written into the Constitution of 1870. In accordance with this plan, the state is divided into fifty-one senatorial districts, from each of which one Senator and three Representatives are elected. The senatorial district is not subdivided into three representative districts, but the three Representatives are elected at large from the entire senatorial district.

In voting for Representatives the elector may mark his ballot so as to cast one vote for each of three candidates, or one and one-half votes for each of two, or three votes for one. Casting three votes for one candidate is called, in the vernacular of Illinois politics, "plumping" the votes. The ballot at the final election contains the names of all the candidates of the various parties and in front of each name is a square. If a cross be marked in the square opposite one name, three votes are to be counted for that candidate; if crosses are marked in the squares in front of two names, one and one-half votes are to be counted for each of the two candidates; if crosses are marked in the squares in front of three names, one vote is to be counted for each of the three candidates.

In practice, the majority party in a senatorial district will usually nominate two candidates and the minority part will nominate one candidate. The electors of the minority party by "plumping" their votes for the one minority candidate, are practically sure of electing him. The electors of the majority party, by casting one and one-half votes for each of the two majority party candidates, are practically certain of electing both. A district would have to show an overwhelming majority in favor of one party to warrant the nomination of three candidates by the majority party. In exceptional

cases, where a district is very evenly divided and may swing in either direction, each of the two major parties would be justified in nominating two candidates; but districts where such conditions prevail are exceptional. Inasmuch as three Representatives are to be elected from each district, the two candidates nominated by the majority party and the one candidate nominated by the minority party are practically sure of election. The final election thereby becomes perfunctory. The nomination is equal to an election. This is one of the defects of this minority system and is a potent cause of bad govern

ment.

This principle of minority representation seems to have solved the problem for which it was devised. For many years past there have been Democratic districts in the northern half of the state and Republican districts in the southern half of the state. This minority system has outlived its original usefulness and remains in our political system, like the vermiform appendix of the human body, a cause of disturbance and disorder.

The Supreme Court of Illinois has held that a primary election is an election, as that term is used in the Constitution, and must therefore conform to the provisions of the Constitution respecting elections. Inasmuch as the Constitution provides that the cumulative plan of voting must be provided in the election of Representatives for the General Assembly, the Supreme Court, by an extension of its logic has ruled that a primary election must likewise provide for cumulative voting. On the face of it cumulative voting at a primary election is absurd, and in practical operation the results produced are disastrous to good government. Inasmuch, however, as the Constitution and the Supreme Court are inexorable on this point, cumulative voting must be provided in our direct primary laws, unless the people shall amend the Constitution to eliminate the principle of minority representation or the cumulative voting feature.

The Illinois legislature has to date enacted four direct primary laws: Law No. 1 was enacted in the regular session of 1905; law No. 2 was enacted at the special session of 1906, called for that sole purpose; law No. 3 was enacted at the adjourned session of 1907-8 following a recess at the conclusion of the regular session of 1907; law No. 4 was enacted at the special session of 1910 called for that and other purposes.

Laws No. 1, 2 and 3 have been declared invalid by the Supreme Court on the ground that they violated provisions of the Constitu