189C.] 40o [Cushins,'. 



the new was in great quantity throughout tlie fields. At the distance of 

 half a league to a league off were large towns, all of them surrounded 

 with stockades. Where the Governor stayed was a great lake near to 

 the enclosure, and the water entered a ditch that wellnigh went round 

 the town. From the River Grande to the lake was a canal, through 

 which the fish came into it, and where the chief kept them for his eating 

 and pastime. With nets that were found in the place, as many were 

 taken as need required ; and however much might be the casting there 

 were never any lack of them." 



Now since the very origin of key building was directly related, in all 

 probability, to the improving of natural, then the making of artificial 

 bayous to serve as fish-pounds ; to the building of fishing stations near 

 by, and resultantly, to the construction of shell settlements in place 

 thereof, we cannot reasonably suppose that the key builders derived 

 all this from the mainland, but rather that the dwellers in the interior 

 here spoken of by an eye-witness, had derived their practice of making 

 such fish canals and preserves, from them or from ancestors like them.* 



If, then, the key-dweller and Southern seashore and flood-laud phase 

 of life and art was, as is here indicated, the originative, the earlier 

 phase, and the mound-builder phase was the later or to some extent 

 inherited phase, it does not follow that the mound builders acquired 

 their art and culture from the particv,lar key dwellers the remains of 



* To state my opinion clearly in reference to this question of the relation of the mound 

 builders to the particular key builders the remains of whom we investigated, I may say 

 that I do not believe this relation to have been necessarily direct, however much it may 

 seem to have been so. The remarkable correspondence in the art characteristics of the 

 mound remains proper, when compared with those exhibited in objects of our collec- 

 tions from the keys of the farther south, signifies to my mind, primarily, that the art 

 displayed in objects from the inland mounds was inherited or derived from key-dwel- 

 lingor sea-dwelling methods of technique and art treatment. This (leaving out all other 

 questions) is indicated by numerous examples of mound art. I need mention only two or 

 three. One is exemplified in the double-bladed battle-axe type of war club, figured in PI. 

 XXXV (3, a). The club of this type that we discovered at Marco was wholly of wood, 

 yet it was evidently, as I have hitherto stated, a survival of the double, semi-circularly 

 bladed war-axe of an earlier time. But it was, nevertheless, a practical, not merely a 

 ceremonial, weapon. Now such a weapon is represented on the embossed copper plates 

 and is engraved repeatedly on the shell gorgets of the mounds, as held in the hands of 

 purely ceremonial figures. It is also sometimes found represented ( among mound-remains, 

 but not among those of the keys) in the shape of small amulets wrought of shell or stone. 

 Again, a single nearly full-sized specimen, made wholly of stone, rather than of wood, (it 

 is beautifully fashioned from light colored flint by chipping and battering, then grind- 

 ing and polishing) has been very recently secured, I understand, by that fine authority 

 on mound archteology, General Gates P. Thruston, President of the Tennessee Historical 

 Society of Nashville, Tenn. All of these mound forms of the weapon, however, are 

 strictly ceremonial ; that is to say they are not directly originative forms, but forms of 

 the weapon inherited and ancestrally venerated, that is, derived from some older form 

 still adapted to practical use— as was the specimen we recovered from Marco. The same 

 may be said of the shapely carving in green-stone, of a nearly full-sized, hafted celt — 

 found in a sepulchral mound in the Cumberland Valley near Nashville, Tenn., some 

 years since, by Prof. Joseph Jones— the correspondence of which as a type form, to 

 the actual celt, found by us at Key Marco, is almost exact, save in merely decorative 

 details of the handle. 



