Per cent

inorganic material.

163,000

.0292

.0102

.0190

65.0

6241.

3.50

1,697

133

3,460

6243.

.0605

.0160

76.0

1

2.75

1,705

133

2,130

.0217

78.8

6253.

.0156

.0030

.0126

80.7

4.480

.0193

.0045

.0148

75.6

.0288

68.2

.0122

.0171

58.4

72,700

.0422

.0032

.0370

87.7

.3620

.0000

.3620

100.0

that the great majority of them were less than 1 or 2 microns, or very much smaller in size than a red blood corpuscle.

"These investigations were sufficient to indicate that it is the very fine particles which are the most important factor in the causation of the dis

deceased miners, by a process of acid oxidation, consisted of extremely minute particles. On measurement, it was ascertained that none of the particles were larger than 12 microns in diameter, and

ease. It is, therefore, very probable that particles over 12 microns in diameter are relatively unimportant, and that only those lying under the limit of 12 microns need be taken into account."

In the light of these conclusions it is evident that the particles which in our study are of greatest interest as disease-producing

ones are those classified as of a standard unit and less in area. One-fourth of a standard unit is equal to 0.0001 square millimeters, or 0.01 × 0.01 millimeters. These particles are, therefore, 10 microns and under in diameter down to, perhaps, 1 or 2 microns, a size which is just countable. It is with these particles that our conclusions will mainly deal. In Table 3 have been grouped the dust counts of 1 standard unit particles and the weight of the inorganic dust per cubic foot of air sampled, with the corresponding Utube readings.

TABLE 3.-Comparative tabulation of number of small dust particles and weight of dust in air with high and low suction heads.

Figure 4 is a graphical representation of the various dust counts and U-tube readings taken from Table 3.

An examination of Table 3 and figure 4 shows that in general, and other conditions being equal, a reduction in suction head is quickly followed by an increase in air dustiness.

The dust counts with low exhaust pressure are relatively high, varying, with two exceptions, from 165,000 to 2,980,000 particles per cubic foot of air. These two exceptions, samples 6252 and 7164, may be justly eliminated. The dust count on these, being less than 50,000 particles at low exhaust pressure, indicates that, relatively speaking, little dust was being created in the operation. It will be noted that while sampling 6252 the operator only worked 10 minutes of the sampling time, and in the case of sample 7164 the operation was very light. The average of all the dust counts is 853,000 particles, and with the two samples 6252 and 7164 omitted, the average becomes 990,000 or practically 1,000,000 particles per cubic foot of air. With the application of the normal, higher exhaust pressure the dust content is considerably lower, an average of all the samples being 200,000 particles. In this case there are again two samples, 6261 and 7161, which greatly increase the value of the average. Disregarding these two values the general average becomes 120,000 particles per cubic foot.

The total weight of total solids per cubic foot of air is 0.069 milligrams with the low exhaust pressure (averaging 0.56 inch) and 0.023 milligrams with the high exhaust pressure (averaging 3.21 inches).

An examination of the corresponding observations in the two halves of Table 3 (the normal conditions being on the left, the reduced exhaust velocities on the right) shows more clearly than the general averages just what was happening. In one case (samples 7163