Vll

what little reliance Dr. Hutton thinks should heplaced upon the present theories. Dr. Robison,who has treated the subject as a mathema-tician, and also experimentally, speaking ofthe inutility of them, says, “ But the clearestproof is, that arches very rarely fail wheretheir load differs most remarkably from thatwhich theory allows.” See art. Arch , Enc. Brit,sup. 1803, and in confirmation, the sections ofEcclesias. Buil. in these Tracts. Gauthey,speaking of the mathematical deductions fromthe theory of La Hire, very truly observes,“ Les recherches analytiques sont malheureuse-ment fondees sur des hypotheses que l’expe-rience dement journellement.”

Hooke first marked the analogy between a cate-nary and an arch, in a cipher • which decipheredmeans, “ ut pendet continuum flexile, sic stabitcontiguumrigidum inversion.” Dr.David Gregorynext derives “ quali vi arcus muros quibus insistitextra propellit; nempe haec eadem est cum partevis catenam sustinentis; quae secundum horizon-talem trahit,” &c. La Hire * then discoveredwhat is called in this country, Attwood’s Theoryof Arches, and he demonstrated a method of de-termining the thickness of an abutment, which,by some extraordinary perversion, has been

* Mem. de FAcad. 1712, and Traite de Mecanique, 1729.Prop. 123, 124, and 125.