Principe and lower in Santiago than in any other province. This suggests that white blood and lawful marriage may be related phenomena. The following table shows that they vary together through the provinces:

Hence it seems probable that legal marriage in Cuba was more common among the whites than among the colored. The following table raises the probability to a certainty:

Legal marriage was more than thrice as general among the whites as among the colored. But even among the whites it was not much more than half as general as it was in the United States. In two former censuses, those of 1841 and 1861, comparable returns have been made. From them the following table has been compiled:

In both races marriage is less frequent than forty years ago.

In the following table the analysis of the subject by race is extended to the provinces:

When the proportion of married in each province is obtained for the two races separately, it appears that white and black were affected by different influences. The highest proportion of married for the whites was not in Puerto Principe, but in and around the capital, and the proportion in the other provinces, except Santiago, was not much below that in Puerto Principe. In this last province the high relative number of married is due in great measure to the fact that the married among the colored are almost twice as many as in any other province. Matanzas and Puerto Principe furnish a striking contrast. In each the married whites were about one-fifth the total whites, but among the Puerto Principe colored one in eight were married, and in Matanzas only one in thirty-eight.

In the following table the analysis of the prevalence of lawful marriage by race has been extended to the fourteen cities separately reported:

With both races marriage is slightly more common in cities than in the rural districts, but the difference for the whites is greater than for the colored.

The married classified by race and sex.-The following table shows the number of married by race and sex:

This shows that the larger proportion of married among the females already noted (p. 120) was confined almost entirely to the whites, and supports the explanation offered that the excess was due mainly to the immigration of husbands without their wives.

The married classified by race and age.-It has already appeared that marriage was more than three times as common among whites as among colored. The same was true of the people over 15 as appears from the following:

In the following table the per cents are given by provinces for each

Among both races marriage was much more common in Puerto Principe than in any other province. But the difference was more marked among the colored than among the whites. The proportion of married among the colored adults of Puerto Principe was one-half greater than in any other province. But among the white adults of that province the married, while twice as numerous as among the colored, were only about one-seventh more numerous than among the whites in Habana city. Marriage was far more evenly distributed among whites than among colored. In the province where it was most general (Puerto Principe) it was less than one-third more common than among the whites of the adjoining province of Santiago, where the proportion of married whites was lowest. But among the colored adults of Puerto Principe marriage was almost five times as common as among the colored adults of Matanzas. The table shows that the local influences favorable to marriage differed widely for the two races. To show this more clearly the provinces may be arranged in the order of the prevalence of marriage among white and among colored adults as follows:

Provinces arranged in the order of increasing prevalence of marriage among—

No relation whatever can be discerned between these two series. Why should Santiago have few marriages among whites and many

among colored, or Matanzas few among colored and many among whites?

The married classified by race and sex and age. In the following table the proportion of married in the adult population of each sex and race is stated.

In preceding paragraphs it has appeared that the proportion of married women in Cuba was slightly greater than the proportion of married men (p. 120), and that this difference was confined to the whites (p. 126); that the difference between the sexes for the total population was less than 1 per cent, but for the whites alone was nearly 2 per cent. The last table shows that for white adults the difference between the two sexes was nearly 5 per cent.

In the following table the facts are given in the same way, by sex and race for the several provinces.

Per cent of adult population (15+) of sex and race specified who were married.

From the preceding table it appears that among white adults the proportion of married females was greater in each province than the proportion of married males. Among colored adults in five of the seven provinces the reverse was true. The difference is due to the excess of males among white adults and of females among colored adults. In all monogamous countries, if either sex is decidedly in the minority, it is almost sure to have a larger proportion of married than the sex which outnumbers it. Among the white adults of Cuba 54.9 per cent were male, but among the colored adults only 47.5 per cent were male. Hence the chance of marriage in the one race is greater among females, and in the other greater among males. This excess of males among white adults appeared in every province of Cuba, and accordingly the higher proportion of married among females was equally general. The excess of females among colored adults appeared in five of the seven