35 



cross breaking- of stringers is concerned, but that the proper units for 

 shearing and for bearing values across grain are either not known or 

 entirely disregarded in a great many cases. Similarly the posts are 

 apparently designed out of reason, but it will be noted that the} T are 

 invariably too heavy, which fact is traceable in part to certain prac- 

 tical reasons to be mentioned later. 



The statement of the author of the paper that "although the string- 

 ers in cross breaking have a factor of safety of 5, and the po.sts have 

 a factor of safety of 20, the structure as a whole has a factor of safety 

 of only 2 approximately " is not warranted as to the conclusion reached, 

 with the whole meaning apparently thereby conveyed. The meaning 

 would seem to be that some vital part of the .structure has a factor of 

 safet} T of 2 only, and hence the structure would be dangerous if loaded to 

 twice the assumed load. As mentioned above, the low factors of safety, 

 according to Table III, are invariably for bearing values. Timber may 

 indent badly, due to an excessive bearing strain across grain, but this 

 would not necessaril} r prove disastrous. It is more a serious question 

 of maintenance expenses than of absolute danger to the structure. 



It is hardly correct, therefore, to conclude, owing to the small bear- 

 ing surfaces at the ends of stringers or of the caps on the posts, that 

 the prevailing practice of timber-trestle construction is subject to as 

 low a factor of safety as 2 in one of its principal parts absolutely affect- 

 ing the safety of the structure. 



As mentioned above, Table IV contains the most valuable informa- 

 tion to the profession at large presented in the paper. It gives the 

 results of the very valuable United States Government tests, mentioned 

 above, corrected, adjusted, and supplemented b} 7 the author of the 

 paper, whose professional standing and work in connection with these 

 Government tests qualify him to present this information in the most 

 authentic manner. The profession owes thanks to Mr. Johnson for com- 

 piling this table with the accompanying explanatory remarks. It would 

 have, perhaps, made the table more valuable to have given extreme 

 breaking unit stresses in place of safe unit stresses, leaving it to each 

 individual designer to select his own factors of safet} T , w T hile a general 

 recommendation as to the desirable factors of safety could have been 

 added.* 



Table V is also of great practical benefit, as it is based on the latest 

 and most reliable coefficients for the cross-breaking stress of timber. 



Table VI would be of even more value to the profession than Table V, 

 provided more information were presented relative to the new formula 

 for columns presented in the paper upon which the table is based. It 

 would be desirable to know the actual number and kind of tests from the 

 results of which the new formula was developed. Presumably the 

 information will be supplied in a future publication of the Forestry 



* They can be obtained by multiplying the safe values by the factors of safety given. 



