GEOLOGY OF THE MERRIMACK DISTRICT. 573 



corner of Concord. The rocks seen on the river road through the town 

 between Concord and Hooksett may be described in the same language. 

 In the south part of Dunbarton and Bow the successive portions will 

 run along the course of Section III. The dips are N. 60'^ W. south of 

 Dunbarton centre, as far as Kimball pond, where there is a change to the 

 south-east, in a ferruginous rock. The south-east dip ensuing continues 

 through the rest of Dunbarton and Bow to Hooksett, to the edge of the 

 Lake gneiss. The strata dip so high that there is room for one or more 

 inversions, and there seems to be a sharp synclinal fold on the east line 

 of Bow. I have noted a possible N. 80"" W. dip near T, Johnson's, in 

 Dunbarton, south-east from Kimball's pond, in a long ridge of bare rock. 

 It is followed a mile south by a similarly inclined quartz. 



Small areas of Montalban rocks occur also in Antrim and Warner, 

 together with the ferruginous tracts in New Ipswich and Deering. 



/\. Ferruginous Schists. 



Three of the designations mentioned at the outset are highly ferrugi- 

 nous : first, portions of the porphyritic gneiss ; second, of the Montalban ; 

 third, the Rockingham mica schist. These rocks may carry six or seven 

 per cent, of red peroxide of iron, and impart a rusty color to all the 

 ledges and the soil for hundreds of square miles. The basis of the rock is 

 twofold : sometimes a nearly pure gray quartzite, full of pyrites, which by 

 decomposition displays to us the rusty colored ledges ; and, again, a mica 

 schist, coarsely grained, probably containing the same compound of iron. 

 Iron is not a mineral that would enable us to use the fact of the presence 

 of a small percentage of it as a satisfactory criterion of geological classifi- 

 cation. But in exploring, one finds it nearly impossible to separate the 

 several ferruginous areas from one another. The ledges rarely show them- 

 selves free from decomposition. The region is not attractiv^e through 

 the great fertility of the soil, and one is inclined to dispose of any question 

 concerning them in the most rapid way possible. The large area in the 

 south-west part of Hillsborough county may be set down as Montalban, 

 and it is represented upon the map by some appropriate modification of 

 that color. It occupies an equally important area in Cheshire county, and 

 is there described as belonging to the Montalban series, page 489. The 

 rocks of the Montalban group in Bridgewater, Bristol, Alexandria, etc., 



