85 



more water it collects aud the inor(i rapidly it wideus at the top and tlio 

 larger the sink becomes. In this way the sinks develop at the same time 

 that the subterranean channels do and in a region of mature sink topog- 

 raphy v.here the channels are Avell below the surface, as in the Indiana 



Fig. 3. A more advanced stage of subterranean drainage than No. 1. Sinks 

 have developed and all the water of the stream passes beneath the surface and 

 enters the larger stream as a great spring. It may be considered a sort of vertical 

 self capture, a common occurrence. The underground channels have become enlarged 

 and subterranean drainage has worked headward along the stream. At this stage 

 the sinks will be developed over considerable of the surrounding land surface. It 

 may be regarded as approaching maturity. 



region, probably ninety-five per cent, of the sinks are formed this way."^ It 

 is certainly true of the Bloomington region and all the Indiana region as 

 far south as Wyandotte that has come under the writer's notice. In some 

 cases a sinli may cover many acres and be as mucli as a hundred feet 

 deep.^ The first surface indication of the sink is frequently the collapse 

 of the soil into the funnel which has been dissolved in the surface of the 

 underlying rocli. This has probably given rise to the popular notion that 

 sinlvs are usually formed by the collapse of the roofs of caverns.' Incipient 



Fig. 4. An ideal section, similar to Nos. 1 and 2, in old age. Natural bridges 

 are developed, much of the roof of the subterranean channel has collapsed, revealing 

 the underground stream, and the mouth of the cavern has retreated by collapse and 

 erosion. Dotted line indicates the old land surface. 



= See Blatchley, 21 Ann. Rep. Ind. Dept. Geol. Nat. Res., p. 133, 1S96. 



° For a discussion of the solution of the Indiana limestones, see Cumings, Proc. 

 this Acad., 1905, pp. 85-102, 1900. raves. F. C. Greene, idem, for 1908, pp. 175-183, 

 1909. 



^ This does not seem to apply to the eaves of Florida aud some regions of Cuba 

 where the channels are very near the surface and the roof soon becomes so weakened 

 that it gives way, and the extreme porosity of the rocks does not concentrate the 

 solution to the joints to produce solution sinks. 



