of a still more complex character. Take, for instance, the letter pp in Landa's alphabet, 
: here are evidently the traces of a face. The same appear, but not so plainly, in the sign for x, which is 
. Now, if we turn to the ancient hieroglyphics upon the monuments of Central America, we will find the human face appearing in a great many of them, as in the following, which we copy from the Tablet of the Cross at Palenque. We take the hieroglyphs from the left-hand side of the inscription. Here it will be seen that, out of seven hieroglyphical figures, six contain human faces. And we find that in the whole inscription of the Tablet of the Cross there are 33 figures out of 108 that are made up in part of the human countenance.
We can see, therefore, in the Landa alphabet a tendency to simplification. And this is what we would naturally expect. When the emblems—which were probably first intended for religions inscriptions, where they could be slowly and carefully elaborated—were placed in the hands of a busy, active, commercial people, such as were the Atlanteans, and afterward the Phœnicians, men with whom time was valuable, the natural tendency would be to simplify and condense them; and when the original meaning of the picture was lost, they would naturally slur it, as wo find in the letters pp and x of the Maya alphabet, where the figure of the human face remains only in rude lines.
The same tendency is plainly shown in the two forms of the letter h, as given in Landa's alphabet; the original form is more elaborate than the variation of it. The original form is 
. The variation is given as 
. Now let us suppose this simplification to be carried a step farther: we have seen the upper and lower parts of the first form shrink into a smaller and less elaborate shape; let us imagine that the same tendency does away with them altogether; we would then have the
