viii PREFACE. 



6th. We learn from the figures in the Manuscript that the cross in some 

 of its forms was in use among this people as a religious emblem, and also 

 that the bird was in some cases brought into connection with it, as at 

 Palenque. 



7th. In regard to the written characters I have reached the following 

 conclusions: 



That, although the movement of the figures is from the right to the 

 left, and the plates should be taken in this way, at least by pairs, yet, as a 

 general rule, the characters are in columns, to be read from the top down- 

 wards, columns following each other from left to right; that when they are 

 in lines they are to be read from left to right and by lines from the top 

 downwards, but that lines are used only where it is not convenient to place 

 the characters in columns. The correctness of this conclusion is, I think, 

 susceptible of demonstration by what is found in the Manuscript. 



8th. That there is no fixed rule in reference to the arrangement of the 

 parts of compound characters. The few which I have been able to decipher 

 sati.sfactorily appear to have the parts generally arranged in an order nearly 

 or quite the reverse of that in which the characters themselves are placed. 



9th. That the characters, while to a certain extent phonetic, are not 

 true alphabetic signs, but syllabic. Nor will even this definition hold true 

 of them all, as some appear to be ideographic and others simply abbrevi- 

 ated pictorial representations. Most of the characters are compound, and 

 the parts more or less abbreviated, and, as the writing is certainly the work 

 of the priests, we may correctly term it hieratic. 



Landa's alphabet, I think, is the result of an attempt on his part to pick 

 out of the compound characters their simple elements, which he erroneously 

 supposed represented letters. The day characters are found in the Manu- 

 script substantially as given by this author, but appear to have been derived 

 from an earlier age, and to have lost in part their original signification. No 

 month characters are found in this work, though common in the Di-esden 

 Codex. 



10th. That the work (the original, if the one now in existence be a 

 copy) was probably written about the middle or latter half of the fourteenth 

 century. This conclusion is reached first, from internal evidence alone; 



