REVIEWS 661 
which their existence is known. The throw of these minor faults varies 
from a few feet to several hundred. The faults enumerated are more fre- 
quent in the trap than in the shale and sandstone. Since faults are 
much more easily detected where they affect the trap, owing to the fact 
that this formation has come to assume the form of ridges since the 
faulting, it is inferred that minor faults affecting the more homogeneous 
portions of the sedimentary part of the system may have escaped 
observation. 
But for the faults, the determination of the thickness of the system 
would be an easy matter. Allowance can be made for the faults which 
are known, but there is no way of taking quantitative account of those 
which have not been discovered. Impressed with the fact that there 
may be many undiscovered faults of slight extent in the homogeneous 
shales and sandstones, Dr. Kiimmel has revised his estimate of the 
thickness of the system, and now gives the following figures: 
Brunswick series, - - - - - 6,000 to 8,000 
Lockatong series, - - - - 3,500 to 3,600 
Stockton series, - - - - - 2,300 to ~ 3,100 
.I11,800 to 14,700 
The thicknesses of the principal sheets of trap are also given, the 
thickest being less than one thousand feet. A brief discussion of the 
conditions of the origin of the system is followed by a résumé of its 
economic resources. 
The report, as a whole, presents in excellent form the results of a 
piece of work which is not likely to need revision. It is a matter of 
congratulation that Dr. Kiimmel has this year been able to extend his 
studies over the Triassic area of New York. 
IR IDES 
The Geological History of the Isthmus of Panama and Portions of 
Costa Rica. By ROBERT T. Hirt, Bulletin of the Museum 
of Comparative Zoodlogy at Harvard College, Vol. XXVIII, 
No. 5, pp. 151-285. Nineteen plates. Cambridge, 1898. 
The results embodied in this report are based on a reconnaissance 
made by the author for Alexander Agassiz in 1895. In spite of the 
fact that the region concerned was only reconnoitered, the report adds 
