data, the questionnaire specifically requested opinion data where actual was not available, and this tabulation, therefore, to a considerable extent reflects ideas rather than tested facts. However, owing to the lack of definiteness concerning many of the more important variables, as outlined above, this tabulation probably is as accurate as is possible to secure at this time.

It is to be noted that Table I concerns untreated ties only. It is also limited to returns which indicated that the failures were more than half caused by decay, and it is, therefore, an effort to compare the decay resisting qualities of untreated timber.

Mechanical Strength.

Cross-ties, as regards "design," must be considered from a somewhat different angle than most parts of a railroad. A tie could, theoretically at least, be designed of such size and strength as to carry the stresses imposed at any given location under given conditions. If the tie were to substantially retain its strength throughout its life, it would be comparable, as regards designing, with rails, bridges, etc. It does, how.ever, lose its strength throughout its life, though we have no data as to whether this loss is proportional to the time (straight line) or otherwise, and it is economy, therefore, to use ties having an initial strength considerably greater than the stresses require, or in other words the new tie has reserve strength. It is believed that the amount of this reserve strength largely determines the life.

A tie performs three principal functions:

(a) Acts as a beam to distribute the load,

(b) Acts as a block to transmit the load from rail to ballast,

(c) Holds the rails in place.

It is, therefore, required to possess beam strength, crushing strength and ability to hold fastenings. As between the three, the beam strength is fundamental, because, if deficient in any given tie, there is no remedy, whereas a deficiency in crushing strength or ability to stand compression across the grain can be remedied by suitable tie plates and the ability to hold fastenings modified by the design of the fastenings.

As between different kinds of timber, the strength can be said to vary with the specific gravity. Table II has been prepared from data contained in Bulletin 556, U. S. Department of Agriculture (Forest Service). The modulus of rupture is used by your Committee as indicating relative beam strength for any given cross-section of tie and the compression perpendicular to the grain at elastic limit as indicative of relative bearing strengths. Diagrams A and B show graphically how the modulus of rupture and the compression at elastic limit vary with the specific gravity. It is to be noted that the modulus of rupture varies about as the first power of the specific gravity and that the compression varies, for tie timbers selected, about as the 2.6 power.

Under the assumptions that the mechanical life of a tie is proportional to its strength, which seems reasonable in view of the surplus strength when new feature, and that the strength is proportional to the specific gravity, Table III has been prepared. Another assumption made in preparing this table is that the relative resistance of various woods to decay is constant for variations in climate.

Effect of Size and Tie Plates.

The replies to the Committee's questionnaire indicated a general belief that ties 7 inches by 9 inches would outlast ones 6 inches by 8 inches when subjected to the same conditions. The replies would, perhaps, average 25 per cent. The data given in the questionnaire relative to ties failing by decay shows an average of 30 per cent. greater life for the 7 inch ties, plated, over the 6 inch ties and 20 per cent. for the 7 inch ties, unplated, over the 6 inch ties. These check the opinions fairly well. It is also found that the value of the tie plate in the case of 6 inch ties is 12 per cent., and the case of the 7 inch ties 21 per cent. All of these percentages must be considered only approximate, due to the character of the data. The greater value of the tie plate in the case of 7 inch tie may be ascribed to the fact that the 7 inch ties are, in general, used under heavier traffic.

Annual Cost.

Diagrams C and D have been prepared from the formula recommended by this Committee for ascertaining annual cost. They are self-explanatory. Applications of Data.

The following examples will indicate how the information contained in this report may be applied:

EXAMPLE 1. In a location where white oak ties costing $1.50 in track have been in use, and where they have failed largely through mechanical wear in an average of six years, how much less per year will chestnut ties at $1.25 in track cost?

more or

Referring to diagram C, it is found that the white oak ties are it is found in column six that chestnut ties have a mechanical costing 301⁄2 cents per year (6 per cent.). Referring to Table III, life compared with the white oak as 50 is to 75, or four years. As per diagram C, this would result in a cost of about 36 cents per

year.

EXAMPLE 2. In a location where white oak ties exposed to light traffic fail by decay in six years, and where the traffic is such that failures are about equally divided, decay and mehanical, cedar ties costing $1.25 in track have been in use.

How much could be

paid for Douglas fir ties, in order that the cost per year would be the same as for the cedar ties? According to column 10 of Table III the cedar ties should average 69 or 5.5 years. As per diagram

D (7

Douglas fir, column 10, gives an average life of 51 or 4.1 years. per cent.) the cost per year is 28 cents, and Table III for As per diagram D, the Douglas fir ties should cost about 97 cents each, in place.

Conclusions.

Tie life data does not exist from which satisfactory conclusions may be reached on the subjects assigned. Information lacking is partly as

follows:

(a)

Actual life data with the conditions of use recorded

completely enough to permit satisfactory comparisons between results on different railroads.

(b) Information as to how the "average" life reported is

obtained.

(c) Information as to whether the relative resistance to of various woods is independent of climate.

decay

(d) Information as to whether the best decay resisting tie untreated is also the best decay resisting tie when treated, whether treatment results in the addition of a uniform number of years life to all kinds of timber, or more to some kinds than others, or whether it results in giving all kinds of ties the same average life under the same conditions.

(e) How mechanical strength is related to resistance to mechanical wear.

(f) How mechanical strength varies with time in track, i. e., from the condition new to the condition when decay necessitates removal.

Recommendations.

(1) The inclusion of the standard test tie in all test installations, for the purpose of comparison, as more particularly outlined above.

(2) The adoption of white oak ties, grade 3, (6′′ x 8′′) class U, untreated, 8 feet long as standard for test purposes.

(3) Laboratory test of the mechanical properties of tie timber, which has been subjected to contact with the soil for varying periods under conditions similar to those to which ties are subjected in track. Such tests should develop the relationship between strength and time exposed to the tie destroying agencies other than traffic.

72 Jack pine.

64 West.yellow pine 85 Spruce.

fo. yellow pine. 74 West.yellow pine 64 Spruce.

68 Spruce.

64 Larch.

80 Loblolly pine.

36

Cherry.

90

Gum

64 Tamarack.

Gum

Mulberry

62 So. yellow pine.

R. & B. oak.

Hackberry.

So, yellow pine. 77 (LL)

West.yellow pine 62 Maple.

Heart, 120-Sap, 30

TABLE II-RELATION BETWEEN SPECIFIC GRAVITY, BEAM STRENGTH AND RAIL BEARING STRENGTH OF TIMBERS USED FOR CROSS-TIES