gauge observations for the period of 8 years (January 1, 1869, to December 31, 1876), at Quincy and Dubuque (normal parts of the river), with those at Saint Louis and Rock Island.

As in the preceding case the Saint Louis gauge-readings have been simply divided by 2, and the midway point of all made common on the diagram (diagram K). This diagram itself shows everything that can be described in words, and only a few points will be noted. First, we see that the stand of the gauge at Rock Island, near mid-height during the months of January and February, is anomalous and does not agree with places above and below it, and must be due to local ice gorgings, and that a great uniformity of relative conditions during these months exists at Dubuque, Quincy, and Saint Louis, and generally high water and low water exist at the same time all along the valley, which is a most useful condition for navigation. Second, we see that the high water in April and May occur at the uppermost place first and succeed in order of time regularly downward, showing that this is mainly what is locally known as a "head rise." Third, the June and July rise seems all the way to partake of the character of a head rise, reinforcing a preceding local one and lengthening its period.

There is an exception in the case at Saint Louis, where an intermediate swell of water in excess appears to come from the Missouri.

The September rise occurs all along the valley at nearly the same time, and is consequently due to general rains throughout the region in this month.

One of the increasing advantages of the low-water navigation as we go lower down the river is shown by the curves in August and September, when there is so much business to do. The surface keeps comparatively up at Saint Louis much more than above.

We have added zone Diagram L for Quincy and zone Diagram M for Dubuque for the 8 years similar to those for Saint Louis and Rock Island for 16 years already described.

The gauge observations at other places which we have, do not furnish means for sufficient periods of time to be valuable as such. They will, however, be considered further on in discussing the question of headway.

HEADWAY UNDER BRIDGES.*

In order to consider the question of headway under bridges we have arranged the data from gauge-readings into tables.

t

1. Saint Louis tables, &c.—At Saint Louis we take the number of days when the river in each month stood as follows: between zero (low water) and 5 feet, between 5 and 10 feet, and so on. This interval, 5 feet, was taken by the board of engineers on the Saint Louis bridge (see Annual Report of Chief of Engineers for 1874, pp. 645-648), and we have followed it, because we could thus continue the same comparisons, and make available the data there given. That data embraces the period from 1861 to 1873, and we have added to it the data for the years 1874, '75, '76. All these data are given in this report in Table A, year by year; the intervals of 5 feet are placed in the first left-hand column; the months at the heads of the columns, and the number of days in the body of the column. From it we see, for example, that in the month of January, 1861, there were 31 days when the river was between zero (low water)

* The tables in the report of the board show the number of days the river was not below a given stage; but the consolidated Table H of the report does not show anything about the periods of head way under the bridge, and is therefore very misleading, owing to the errors in its construction.