[319-1] This is one of the exceptional cases.

[321-1] Second exception.

[321-2] Third exception.

[323-1] One line has been omitted in the numeral symbol.

[323-2] Here we have again the added day.

[324-1] The 8 at this point in the codex is an evident error.

[324-2] Here is also an error in the original, this being 10.

[324-3] The symbols require an additional day here.

[324-4] The 8 in the year line in the original is a manifest error, as 6 precedes and 7 follows.

[324-5] The 18 in the day line at this point is also an error, as the interval between 2 Muluc and 10 Cimi is 8 months and 17 days. Moreover, the next day number being 16 requires this to be 19.

[325-1] The counters in the original at this point are certainly wrong, for here should be 7 months and 8 days, whereas the symbols are those for 8 months and 17 days.

[325-2] Here we have again the additional day.

[325-3] Added to show connection with the lower series.

[333-1] Codex has 19, which is equivalent to 1 year and 1 month.

[337-1] While reading the final proof I fortunately discovered what may prove to be the correct explanation of the numbers in the loops.

At the commencement of the series on Plate 71 and at its close on Plate 73 we observe the symbol of the day, 9 Ix. Starting from this date and counting forward on the calendar two months and fourteen days, we reach 11 Lamat. This gives the number in the first loop of the series. Two months and fourteen days more bring us to 13 Ik, the number in the second loop; two months and fourteen days to 2 Cib, the number in the third loop, and so on to the end. It is therefore probable that the numerals in the loops indicate the week numbers of the days, though these are usually expressed in red symbols.

[338-1] The 7 in the twelfth column is an error; it should be 8, as an inspection shows the place of the missing dot. The additions make it clear that the numbers of the second line refer to months, those of the line below them to days, and those of the line above to years. The series is, therefore, apparently complete without the numbers inclosed in the loops.



CHAPTER II.

CONCLUSIONS.

The conclusions to be drawn from the foregoing discussion may be briefly stated as follows:

First. That the codex in its present form is composite, being made up from two or more different original manuscripts, as Dr. Foerstemann has suggested.

Second. That a number of minor changes and additions have been made by a subsequent hand, possibly after it had assumed its present form.

Third. That the year referred to in the larger series is one of 360 days; also, that in instances of this kind the count is continuous, and hence not consistent with the generally received idea of the Maya calendar, in which, the four year series forms a necessary part of the system, unless some other method of accounting for the five supplemental days can be discovered than that which has hitherto been accepted.

Fourth. On the other hand, indications of the four year series are certainly found in all of the Maya manuscripts; for example, in Plates 25-28 of the Dresden Codex and Plates XX-XXIII of the Manuscript Troano,[339-1] which seem to be based on this series; in fact, the numbers attached to the days in the latter can be accounted for in no other way. Plates 3-6 of the Cortesian Codex are apparently based upon the same system. The numbers in the loops on Plates 71, 72, and 73, Dresden Codex, heretofore alluded to and represented in Fig. 371, apparently defy explanation on any supposition except that they refer to the numbers of the ahaues, which are based upon the four year series.[339-2] The frequent occurrence in connection and in proper order of both the first and the terminal days of the year apparently refers to the same system. Many of the quadruple series no doubt relate to the four cardinal points and the four seasons; yet there are some which cannot be explained on this theory alone.

It is impossible, therefore, to exclude this system from consideration in studying the chronology of the codices, although there are a number of the numerical series of the Dresden manuscript which cannot be made to fit into it on any hypothesis so far suggested. The same thing is also found to be true in regard to some, in fact most, of the series found in the Mexican manuscripts. This confusion probably arises in part from the apparently well established fact that two methods of counting time prevailed among both Mexicans and Mayas: one, the solar year in ordinary use among the people, which may be termed the vulgar or common calendar; the other, the religious calendar used by the priests alone in arranging their feasts and ceremonies, in which the cycle of 260 days was taken as the basis. But this supposition will not suffice as an explanation of some of the long series of the Dresden Codex, in which the year of 360 days appears to have been taken as a unit of measure, unless we assume—as Foerstemann seems to have done—that what have been taken as years are simply high units and counting the whole as so many days, refer the sum to the cycle of 260 days, which will in almost every case measure them evenly as a whole, or by its leading factor, 13. That the smaller series attached to day columns are all multiples of 13 and referable to the cycle of 260 days has been shown by Foerstemann as well as in the preceding part of this paper. But it is worthy of note that the difficulty mentioned occurs only in reference to series found in that portion of the Dresden manuscript which Foerstemann has designated Codex B (page 24 being considered as belonging thereto).

The red unit number symbol, with a circle of dots around it, seen occasionally in the Manuscript Troano, seems to have some connection with the four year series. Take, for example, the one in the lowest division of Plate VII.

The series commences in the lower right hand corner of Plate VIII, where the day column with which it is connected is found. The days of this column, reading downward, are as follows: Ahau, Eb, Kan, Cib, Lamat, and the number over them is I, but without any dots around it, while the terminal I of the series is inclosed in the circle of dots. What is the meaning of this marked distinction? It is evident that it is something which does not apply equally to all the days of the columns; yet, as it is the terminal number, it must relate to some one of them. If we examine the series carefully I think the reason for the distinction will be explained; Written out in full, it is as follows:

I. Ahau Eb } Kan } 10, XI; 10, VIII; 10, V; 10, II; 12[?], [(I)]. Cib Lamat

The last black number is 10 in Brasseur's fac simile, but should be 12. Making this correction, the series is regular and of the usual form. The sum of the black numbers is 52, which is the interval between the days, and the number over the column is the same as the final red number.

If we turn now to the calendar (Table II) and select Ahau of the Kan column, and 1, the seventeenth number of the eighth figure column, and count 52 days, we reach 1 Eb, the second day of our column as given above; 52 days more bring us to 1 Kan, the first day of the first month in the calendar and third day of our column. If the theory of the four year series be correct, then 1 Kan of the Kan series must be the first day of the first year of an Indication or week of years. This fact was probably considered by the aboriginal artist of sufficient importance to give this day a mark of distinction. As it is not possible for any of the other days of the column to be thus distinguished, it is fair to presume this peculiar marking of the final number refers to Kan. Moreover, this distinction would not occur if any other than the Kan series were used.

In the upper division of Plate IX of the same manuscript is the following series:

XIII Men } Manik } 20, VII; 20 [(I)]; 1, II; 4, VI; 7, XIII. Cauac } Chuen Akbal

In this, I, the second red number of the series, has the circle of dots around it. The number over the column is partially obliterated, but is readily restored, and should be XIII.

If we select, on our calendar, the Cauac column, or series, a reason for this distinction will appear. The sum of the black numbers is 53, which is also the interval between the days. As has heretofore been shown, the red numbers of the series refer to certain days selected by the priests, for special reasons unknown to us, which occur between the days of the column.

In this case the intermediate days are as follows:

Between 13 Manik and 13 Cauac: 7 Manik, 1 Manik, 2 Lamat, and 6 Eb.

Between 13 Cauac and 13 Chuen: 7 Cauac, 1 Cauac, 2 Ahau, and 6 Kan.

Here we find the explanation for which we are seeking, as in the interval between 13 Cauac and 13 Chuen is 1 Cauac, which, if the Cauac column of the calendar be selected, is the first day of the year 1 Cauac, the first year of an Indication. As this occurs only when a year commencing with Cauac is selected, we infer that the series is based upon the system with the four year series.

The best illustration of this peculiarity and the strongest evidence of its signification is probably found in the series contained in the middle division, Plate XI, same manuscript. This, when written out and the numbers properly arranged, is as follows:

[(I)] [(I)] Oc Ahau } Cib Cimi } 1, II; 2, IV; 2, VI; 5, XI; 2, XIII; 4, IV; 9(?) [(I)]. Ik Eb } Lamat Ezanab Ix Kan

The last black number of the series is 9, but should be 10 to render the series complete. Making this correction, the series is of the usual type; the sum of the black numerals is 26, the interval between the days of the columns is 26, and the final red numeral is the same as that over the columns.

As the circle of dots is around the final red number and also around each of those over the columns, the distinction indicated must refer to one or more days of each column.

As the last days only of the columns are year bearers, the mark of distinction probably applies to them. Selecting for the left hand column the Ix series of years and commencing with 1 Oc, the seventeenth day of the eighth month, we count 26 days. This brings us to 1 Cib, the third day of the tenth month, or tenth figure column of our calendar and second day of the first day column of the series; 26 days more to 1 Ik; 26 more to 1 Lamat, and 26 more to 1 Ix, the first day of the year 1 Ix, which, according to the four year series, will be the first year of an Indication. Selecting the Kan series for the second column and counting in the same way from 1 Ahau, the seventeenth day of the eighth month, or eighth figure column of the calendar, the last day is found to be 1 Kan, the first day of the year 1 Kan, which must also be the first year of an Indication.

Unit numerals marked in this manner are found in two or three places in the Cortesian Codex, but there is none in the Dresden Codex. The series with which they are connected in the former, except that in the middle division of Plate 24, are too much obliterated to be traced throughout. This, by making two slight and apparently authorized corrections, is as follows:

[(I)] Cimi } Ezanab } 11, XII(?); 11, X; 6, III; 8, XI; 7(?), V; 9, I. Oc } Ik Ix

The first red numeral of the line is X in the original and the next to the last black number is 6. By changing the former to XII and the latter to 7 the sum of the series will be 52, which is the interval between the days of the column.

Using the Ix column in the calendar and commencing with 1 Cimi, counting as heretofore, the last day of the column of the series is found to be 1 Ix, the first day of the year 1 Ix and the first year of an Indication, according to the four year system.

A somewhat remarkable confirmation of the theory here advanced is presented in a series found in the middle division of Plate II of the Manuscript Troano.

The series, when written out with the substitutes heretofore used, is as follows:

[(I)] [(I)] Manik Ymix } Men (?) Been } 9, X; 6, III; 11, I. Chuen Chicchan } Akbal Caban Men Muluc

In Brasseur's fac simile the second symbol of the left hand column is clearly that for Men. If this be accepted as correct, then no year bearer (Kan, Muluc, Ix, Cauac) would be found in either column and the theory we have advanced regarding the signification of the dots around the red unit over the column would fall to the ground. Nor is this the only difficulty we meet with in attempting to apply the theory to this series. The sum of the black numbers is 26, which should also be the interval between the days of the columns. Counting 26 days from 1 Manik brings us to 1 Been instead of 1 Men; 26 more to 1 Cauac, a day not found in either column as given in the original. Taking the second column and counting 26 days from 1 Ymix, we reach 1 Manik, instead of 1 Been. This gives us the key to the series and solves the riddle. We must commence with 1 Ymix, then take 1 Manik, then 1 Been, and so on, going alternately from column to column.

Adopting this method and using the Cauac column of our calendar, Table II, the result is as follows: Commencing with 1 Ymix, the third day of the tenth figure column, and counting 26 days, we reach 1 Manik; 26 days more bring us to 1 Been, and 26 more to 1 Cauac, the first day of the first year of an Indication. The 1 Men of the left hand column should therefore be 1 Cauac, which is also proved by counting the intervals, without regard to the week numbers. For example, from Ymix to Been is 12 days, from Been to Chicchan 12 days, from Manik to Cauac 12 days, and so on through each column. Or, if we take the columns alternately, the interval is six days, thus: From Ymix to Manik, 6 days; from Manik to Been, 6 days; from Been to Cauac, 6 days; from Cauac to Chuen, 6 days, and so on to the end.

Although the proof is not absolutely conclusive that these red unit numerals have this mark of distinction for the reason given, it nevertheless furnishes what would seem to be a satisfactory explanation, and, if so, affords proof that the calendar system, based upon the four year series, was in vogue when the Manuscript Troano and the Codex Cortesianus were written.

This mark of distinction is found in a strange and unusual relation in the lower division of Plate XV, Manuscript Troano. The first red numeral of the series is given thus:



Most of the day and about half of the numeral symbols are obliterated, but all that are necessary for present purposes remain distinct and uninjured, as follows:

III, } Ix } 10, XI[(I)]I. Cimi }

Judging by these and the few numbers remaining, the entire series was as follows:

III, } Ix } Cimi } Ezanab } 10, XIII; 4, IV; 20, XI; 9, VII; 9, III Oc } Ik }

The only doubt in reference to the restoration is whether the second and third pairs of numerals should be as given, or 2, II, and 22, XI. If we select the Kan column of our Table II and count from 3 Ix of the eleventh figure column, we reach 13 Kan. If the four year series was the system used 13 Kan might be the first day of a year, but not the first day of an Indication. As this is the only day referred to by the XIII which could have been the first of a year we must seek an explanation in something else. Counting ten days from 3 Ezanab will bring us to 13 Lamat, which is the last day (counting the five added days) of an Indication, commencing with the year 1 Kan and ending with the year 13 Kan.

According to my theory of the ahaues,[344-1] the year 13 Kan would have corresponded with the Gregorian years 1376, 1438, 1480, and 1532. According to the theory advanced by Perez,[344-2] it would have corresponded with 1385, 1437, 1489, and 1541.

It is therefore possible that this mark of distinction may be of some value in determining the relation of the Maya to the Gregorian calendar.

FOOTNOTES:

[339-1] See Study of the Manuscript Troano, by Cyrus Thomas.

[339-2] See note on page 337.

[344-1] See Table XVII, Study of the Manuscript Troano, by Cyrus Thomas, p. 44.

[344-2] See Table XVIII, ibid., p. 45.



CHAPTER III

THE WRITING.

It must be admitted that none of the attempts made at deciphering the writing in these manuscripts has proved entirely satisfactory; in fact there is still some doubt as to whether any of the characters are truly phonetic; nevertheless it is believed that what is here shown will tend to lessen this doubt. It must be conceded, however, notwithstanding these drawbacks and difficulties, that some material progress has been made towards a better understanding of its type and of the nature of the characters.

The direction in which it is to be read must of course be determined before any progress can be made in deciphering it. This was, until recently, a matter of speculation, but now may be considered settled. As this has been explained[345-1] it is unnecessary to repeat that explanation here.

A certain parallelism in the sentences or groups of characters has also been discovered. Attention was first called to this by me in the work referred to, but is more fully explained by Dr. P. Schellhas in his paper entitled "Die Mayahandschrift der koeniglichen Bibliothek zu Dresden." It will readily be understood from a single illustration. Take for example the lower division of Plate XV of the Manuscript Troano (see Study Ms. Troano). Omitting from consideration the numerals and the day column at the left, there are here two short columns on the left and two on the right over the animal figures, and three longer columns between. As explained in the work referred to, the short columns are to be read as lines from left to right and the longer columns separately, from the top downward. There are, in all, five groups or sentences, each containing four compound characters. Representing these by letters, repeating those which indicate similar characters, and arranging as in the plate, the result is as follows:

b a h l m w a r n a a a r s r r r p k t

In this case the characters represented by a and r are repeated in each group and in the same relation to the other characters. It is apparent, therefore, that each group is to be read separately, and, as each repeats in part what is given in the others, it is more than probable that they are simply short formulas to be repeated in certain religious ceremonies. This parallelism, though not always so apparent as in the case presented, is nevertheless found running through all the codices. The advantage to the attempts at decipherment which results from this fact is evident, as it will often justify the restoration of blurred or obliterated characters, and, what is of still more importance, will enable the investigator to test his conclusions by comparing the different characters and pictures with which they are associated.

Although it appears to be well settled that, as a rule, the writing, when in lines, is to be read from left to right—the lines following each other downward and the columns to be read from the top downward, but the groups, as before explained, to be read separately—it does not follow that the groups succeed one another from left to right. This has generally been taken for granted, but there are some reasons to doubt the correctness of this conclusion as regards a number of plates and possibly one entire codex.

The facts that the lines of numerals attached to the day columns extend to the right and that the written characters, when in lines, follow one another in the same direction lead us to infer that the groups and pictures follow one another in the same order, but the apparent movement of the latter towards the left would seem to indicate that they follow one another in this direction. This inference appears to be confirmed by the following evidence: As is well known, the plates of the Manuscript Troano are to be taken in reverse order to the paging. Turning to Plate II, we observe in the middle department of the middle division a bound captive or victim, on whose neck a machete is descending to sever the head from the trunk. Turning to Plate III, which properly stands to the left of Plate II, we see a headless trunk covered with blood and the fatal machete near the neck. It is fair to presume that this is the same individual that is figured in the preceding plate, and, if so, that the pictures follow one another toward the left.

Placing Plates XV* and XVI* of the same manuscript in the proper relation to each other and carefully examining the figures in the second division, we notice that the idol heads which the artisans are carving approach completion as we move toward the left, those in Plate XV* and the right hand one in XVI* being simply blocked out, while the middle one in the latter plate is completely rounded and is receiving the second ornamental line and the one at the left hand is receiving the third and final line.

The female figures in the second division of Plate XIX* indicate the same order, as shown by the increasing girth as we proceed toward the left.

The same order appears to be indicated in numerous places by the symbols of the cardinal points inserted in the text, as they (supposing the conclusion as to their assignment in my "Notes on certain Maya and Mexican manuscripts," accepted by Drs. Foerstemann and Schellhas, to be correct) follow one another in the proper order if read towards the left, to wit, south, east, north, west.

As the writing over each figure, consisting usually of four compound characters, appears to refer to that over which it is placed, it follows that these character groups must be taken in the same order as the pictures. The suggestions on this point are presented here more as proper subjects of investigation by students of American paleography than as fixed conclusions of the writer. If found to be justified by the facts, they will furnish some additional aid in the work of deciphering these manuscripts.

SIGNIFICATION OF THE CHARACTERS.

As Landa's alphabet has so far proved useless as an aid in deciphering these manuscripts, our only hope of accomplishing this end is by long and careful study of these records and laborious comparisons of characters and the relations in which they stand to one another and to the figures.

Some discoveries made while preparing this paper for the press, which are mentioned further on, may possibly give us the key to the method used by Landa in forming his alphabet, and, if so, will probably furnish some slight additional aid in our investigations.

The direction in which the writing is to be read having been ascertained, our next step is to determine by comparison the probable signification of as many characters as possible before discussing the question of phoneticism. The relation of the characters to the pictorial representations forms our chief reliance in this branch of the investigation.

As a commencement in this work and as a basis for further attempts in the same direction, attention is now called to some characters, other than the day and month symbols, whose signification seems to be satisfactorily determined. As there is still some difference of opinion as to the assignment of the symbols of the cardinal points they are also omitted from the list. M. Leon de Rosny has given, as a supplement to his edition of the Cortesian Codex, a list of characters with their supposed signification. It is not my intention to discuss here the merits of this vocabulary, although I shall avail myself of so much found therein as appears to warrant acceptance.

The question of phoneticism will not be considered in connection with the list, as the subject will be briefly discussed at the close, the only object in view in giving the list being to indicate the signification of the characters alluded to. The Maya names appended are therefore to be understood simply as the supposed names applied to them or the objects they denote.

SYMBOLS OF ANIMALS &C.



Kal. The symbol for the number 20. Found in all of the codices and explained in the preceding portion of this paper.



The symbol for 0 (nought), always red. Found only in the Dresden Codex and always in the numeral series.



Kin. Sun, and probably day also. It is not known positively that it has this signification except in connection with the equatorial cardinal point symbols and the symbol of the month Yaxkin; yet it is reasonable to suppose it has.



Aac or Ac. A turtle. That this symbol as shown in a and b denotes the turtle is conclusively proved by its resemblance to the head of that animal, as figured in the Cortesian Codex (see Fig. 373) and its relation to these figures. Found only in this codex, unless two doubtful symbols on Plate XXV*, Manuscript Troano, are to be considered as variants.



There can be no doubt that Landa's A, an exact copy of which is given in the margin, in both varieties, c and d, is nothing more nor less than this symbol; for, in addition to the very close general resemblance, we see in it the eye and the dot indicating the nostril. This fact is important, as it gives us some clew to the method adopted by Landa in forming his alphabet.



Uech. Symbol or head of the armadillo of Yucatan. Appears but once or twice and in the Manuscript Troano only. (See Study of the Manuscript Troano, by Cyrus Thomas, pp. 98 and 145).

[TN-10]

Che. Wood. (See Study of the Manuscript Troano, by Cyrus Thomas, p. 144).



Cab. Earth, soil; also honey. (See Study of the Manuscript Troano, by Cyrus Thomas, p. 150.)



Piz. Stone or stone heap. (See Study of the Manuscript Troano, by Cyrus Thomas, p. 144). The Maya name of the thing indicated is uncertain, though I am inclined to believe Piz, as given in the work alluded to, is correct.



U. The left symbol of this figure appears to stand for vase, and is also used to indicate a pronoun or article when joined to another symbol, as here shown. (See op. cit., p. 145.)



Xicim. The ear. Rosny, Vocabulaire hieratique, No. 185.



Hau. The quarter of a deer. Usually represented as an offering to the gods; in all the manuscripts.



Ikilcab. The bee. Although the figure bears a much stronger resemblance to a beetle than to a bee, there can be no longer any doubt that Brasseur's supposition that it represents a bee is correct.



Honey in the comb. (See Study of the Manuscript Troano, by Cyrus Thomas, Fig. 20); in the Manuscript Troano only, and always in red.



Xamach or Chimix. A vessel. This symbol, found in all the codices, is apparently explained by its use in the upper division of Plate 27, Cortesian Codex, where it stands over each of four vessels or jars of the form represented in Fig. 374.



This conclusion is greatly strengthened by the fact that the only other symbols in this connection are those of the cardinal points, one to each vessel. These figures are probably intended to denote here the four sacred vessels or amphorae of the Bacab, though not surmounted, as Brasseur supposed, by human or animal figures.

The symbol appears to be used also in the ordinary sense, or at least to signify other vessels than the sacred four, if we may judge by its frequent repetition in Plate XIV, Manuscript Troano. But it is worthy of notice that here also, in both the middle and lower divisions, four of the symbols are connected with the cardinal point symbols; there is also in the former the figure of a vessel.



If this identification be correct it is important, as it has a strong bearing on the question of phoneticism. It will be observed that, although the right hand member resembles closely the symbol of the day Ymix, there are some differences, as may be seen by comparison. In the former the little figure at the top is divided as in Kan, and on each side of it there is a large dot, usually, and apparently by intention, circular or hollow. These differences are permanent in the different codices.

In the upper division of Plates X and XI, Manuscript Troano, where this symbol appears in connection with each of the four cardinal symbols, that relating to the east presents this remarkable variation:



(?) A conventional figure of sprouting maize, never inserted in the text, but frequently in the Manuscript Troano and in the Peresian Codex made a part of the head gear of figures of deities, in which case the Kan symbol is generally omitted.

The Kan symbol in this connection cannot be intended, as Dr. Schellhas supposes, to indicate the field or milpa in which the corn is growing, but the grain from which the plant is springing. (On this subject see Study of the Manuscript Troano, by Cyrus Thomas, pp. 105 and 107.)



(?) Symbol of a worm which gnawed the roots of the growing agave or maguey; appears but once, on Plate XXIXc of the Manuscript Troano.

The animal head and teeth show the erroneous idea the natives had of the gnawing apparatus of insects. The worm is shown on the next page in Fig. 375.



Chuplal. Woman or female. This symbol is found in the Dresden and Troano Codices, but most frequently in the former. The appendage at the right is sometimes wanting, and occasionally that at the left, but when this is the case some other prefix is generally substituted.

If we examine carefully Plates 16-20 of the Dresden Codex, where this symbol is most frequently repeated, and compare it with the heads of the females there figured, it soon becomes apparent that the scrolls with the heavy black dot are intended to denote the locks of hair and that the symbol as a whole is, as usual, a modified or conventional form of the head (see Fig. 376).



Otoch. A house or dwelling, or Tabay; a hut or hunting lodge. The symbol marked a is found in the Cortesian Codex on Plate 29; that marked b, on Plates 29, 32, and 34, same codex, and on Plates XVI* and XXII* of the Manuscript Troano. The one marked c is the usual form in the latter, as on Plates V*, VII*, and X*. It is also on Plate 38 of the Dresden Codex.

The relation of these symbols to the conventional figures of houses or huts inserted at the points where they are found, together with the form, which shows an attempt to represent the thatched or leaf covered roof, leaves no doubt that they are used for the purpose indicated.



Buk (?). There are good and, it is believed, satisfactory reasons for concluding that these symbols are intended to denote the action of whirling a stick to produce fire or rolling a pestle in grinding paint. The first, marked a, is found only on Plate XIX of the Manuscript Troano, and the second, on Plates 5 and 6 of the Dresden Codex.

A copy of part of Plate XIX of the Manuscript Troano is introduced here (see Fig. 377) to show the relation of the figures to the characters. If this interpretation be correct, we see here an evident attempt on the part of the aboriginal artist to indicate by the symbol the action necessary in the work to be performed. It is probably a conventional sign, and not a phonetic character.



(?) In all probability one of the symbols used to denote the act of walking or taking steps. Found but seldom in this particular form, though each portion occurs frequently alone or in other combinations.



A remarkable series of figures and written characters runs through the lower division of Plates 65 to 69 of the Dresden Codex, apparently devoted entirely to the representation of incidents in the life of the culture hero Kukulcan, or deity mentioned on a subsequent page as the "long nosed god" or "god with the snake-like tongue," or to ceremonies to be performed in honor of this deity. Over the figure are three lines of written characters, as shown in Fig. 378, which is a copy of the lower division of Plate 65. These, as is readily seen, are in groups, one group of six compound characters over each figure of the god. There are thirteen figures of the god and thirteen of these groups of characters in the series. The characters of a group, as may be seen by reference to the figure, are arranged in the following manner:

___ _a_ _b_ __ __ _c_ _d_ __ __ _e_ _f_ __ __

to be read (presumably) in the alphabetic order of the letters given; though the order in which they are to be read is not essential at present. Examining the series carefully we find that the first character of each group corresponding with a in the above diagram is the same throughout. The same thing is true in reference to the third, or that occupying the place of c in the diagram, which is the symbol of the deity. The sixth, or that corresponding with f in the diagram, is also the same throughout the series; the fifth, corresponding with e, is substantially the same throughout, though subject to more variations than any of the other characters. It follows, therefore, that the chief and almost the only differences in the readings of the groups are to be found in the second and fourth characters, or those represented by b and d in the above diagram; the others (at least those represented by a, c, and f), if referring at all to the figures, must relate to something found in or applicable to each. The third (c), as stated, is the symbol of the deity and corresponds in the text with the figure of the god in the pictures. As this deity figure is the only thing found in all of the representations, we must seek for the explanation of the other two permanent characters in something else than what is figured.

Comparing the second character (b) of each group with that upon which the god is seated or standing, we find sufficient evidence to satisfy us that this symbol is the one which is used throughout to indicate this object. For example, the second symbol in the group on Plate 69 is an exact copy of the object on which the deity is seated. The same thing is substantially true of that in the left hand group of Plate 66, the middle group of 67, and the right hand group of 68.

Assuming, on account of the remarkable regularity of this series and the fact that the deity is in each case seated or standing on something, that this rule holds good throughout, we have a clew to those corresponding symbols which are not simple copies of the things they are used to indicate.

Turning to Fig. 378, we observe in the right hand department the marks of footsteps under the deity and the character shown in the margin (No. 20) as the second of the group above the deity. It is worthy of notice that in the two we find precisely Landa's two characters for the letter B. Is it possible that the two principal parts of this compound character denote the Maya words oc be, "foot journey" or "enters upon the journey"? Attention will be called to this further on, but it is proper to state here that as the prefix is found in three other corresponding characters it cannot be a necessary part of that which represents the footsteps in this case.



Assuming the theory above given as to the characters in the inscription which represent the things under the deity figures to be correct, the second character in the middle group of the lower division of Plate 65, shown in Fig. 378, will be the symbol for the substance represented by scrolls under the figure of the deity.[354-1]

The prefix in this case is the same as that to the symbol above described (No. 20), and of course has the same signification. The other portion of No. 21 must therefore represent the substance in which the god is walking. This appears to be dust, sand, or mud.



Cacauak or cacauche. The wild or cultivated cacao. Found a number of times in the Dresden Codex, sometimes as represented in the marginal figure a and sometimes as in c, and always in connection with figures holding in the hand a fruit of some kind. It appears once in the Cortesian Codex (Plate 36), as shown in b, in connection with a fruit of precisely the same kind as that figured in the Dresden Codex. It is found also on Plate XVIII* of the Manuscript Troano, but is apparently used here to denote an action.

There can be little, if any, doubt, judging by the figures in connection with which it is found, that this symbol is used in the Dresden and the Cortesian Codices to denote the cacao. Whether it refers to the tree or to the fruit is uncertain; possibly the different forms in which it is found are intended to denote these distinctions. In some of the figures the capsule appears to be indicated; in others the seed. The prefix to figure c apparently indicates the heaping or piling up of the fruit on the dish held in the hands of the individuals figured in the same connection, as, for example, on Plates 12 and 13 of the Dresden Codex. If this supposition be correct it gives us a key to the signification of this prefix. Reference to its use in the upper division of Plate XVIII*, Manuscript Troano, will be made further on.

In this symbol we find another of Landa's letters, and, if phonetic, agreeing precisely with his interpretation.



Ekbalam according to Rosny. The variety marked a is found twice in the Manuscript Troano, Plates XVI and XVII, and that marked b once in the Dresden Codex, Plate 8, each time in connection with a spotted, leopard-like animal.

The black markings on the symbols render it probable that Rosny's interpretation is correct. The numeral before the first form may possibly be explained by the fact that this symbol is used once (Manuscript Troano, Plate XII) to indicate the day Ix.



Moo. The ara, a large species of parrot. This symbol is found but once, and that in Plate 16c, Dresden Codex, in connection with the bird shown in Fig. 379.



The conclusion in this case is based on the following evidence: In this series there are six groups of characters, four compound characters in each group, arranged as in the annexed diagram:

_________ _a_ _b_ _e_ _d_ _g_ _h_ _i_ _m_ _o_ _c_ _d_ _c_ _f_ _c_ _d_ _c_ _n_ _b_ __ __ __ 1 2 3 _k_ _c_ _c_ _l_ _l_ _p_ __ __ __

Similar characters in the different groups are represented by the same letter; for example, the symbol for woman, heretofore shown (No. 17), is represented by c, and an unknown character by d. Different letters represent different symbols. It is apparent that we have here the parallelism heretofore spoken of and are justified in basing conclusions on this fact.

At 1, 2, and 3 are female figures with a bird in each case perched on the back. At a is the head of a bird, evidently the symbol of the bird on the female below; at i, in the fourth group, is precisely the same symbol as the one found in the same relative position in the middle division of Plate 17 over another bird, and at m, in the fifth group, is another bird's head. From these facts we conclude that the first symbol in each of these groups denotes a bird, and, as no two are alike, that they refer to different species, the one at g corresponding with symbol No. 24, the bird beneath being the great parrot or ara. Other facts, derived from a careful study of the various groups of this portion of the codex, which would require much space and numerous illustrations to explain, lead to the same belief.

According to this conclusion, the following symbols also denote birds, probably of the species here indicated.



Icim? The horned owl. This is represented by a in the first group in the above diagram.

The bird in the figure under the group, although horned, bears but slight resemblance to an owl; yet, comparing the marks on the tail with those of two of the birds on Plate XVIII* of the Manuscript Troano, I think the interpretation is justified.



Kukuitz? The Quetzal. The symbol is apparently incomplete, but the bird figured under it justifies this conclusion. This symbol is represented by e in the above diagram.

If this interpretation be correct, we find in this symbol another of Landa's letters.



Kuch. A vulture or bird of prey much like the sopilote. These two symbols (a and b) appear to refer to the same bird, evidently a vulture. (See Manuscript Troano, Plates XVIIa and XXVI*a.) The first form (a) is found but once (Manuscript Troano, Plate XVIIa), the other at several points, both in the Manuscript Troano and the Dresden Codex, and is represented by m in the preceding diagram.

If this determination be correct, the first of these symbols (a) is probably phonetic and agrees with the interpretation of No. 26.



Chom, Xchom, or Hchom. The sopilote or vulture. Found only in Plates 16 and 17, Dresden Codex. The bird figure in Plate 17 appears to be intended to represent a vulture. The symbol corresponds to i in the preceding diagram.

If phonetic, the word indicated should, according to Landa's alphabet, be aspirated, which is found to be true of one of the forms given by Perez.

In certain series of the Dresden Codex, which appear to relate to the four year series or to the four seasons, especially those on Plates 29-31, a certain class of food animals seems to be assigned to each. The four following symbols are those used to express this idea:



Ceh? The symbol for game quadrupeds. The same idea appears to be indicated by the folded and tied quarter of a deer, as shown in No. 11. The head shown in the symbol is probably intended for that of the deer, though more like that of the rabbit.



Cutz or Cax. The symbol for game birds, the head being probably that of the wild turkey (Cutz or Ahcutz).



Huh. The symbol for food reptiles or the iguana.

As the Kan figure is admitted to be a maize or bread symbol, it is readily seen that the object in view in connecting it with the animal figures is to indicate that they are used for food, and hence are proper offerings to the gods, which is equivalent to saying, to the priests.



Cay. The symbol for food fishes, or fishes in general, though as often on the Kan symbol or without any suffix.



Cutz or Cax. In one of the two series of these food symbols, in Plates 29-31 of the Dresden Codex, in place of the bird symbol No. 30 is that shown in symbol No. 33. It probably has, as Rosny supposes, the same signification, a supposition which is strengthened by the fact that it is found in the bird series on Plates 16c and 17c, same codex, and is represented by o in the preceding diagram.

SYMBOLS OF DEITIES.



Ekchuah. The symbol or hieroglyph of the deity named "Ekchuah" by the Mayas and considered the patron and protector of peddlers or traveling merchants (Fig. 380).



The signification of the name of this deity is "The Black Calabash." The form and the shading of the symbol render it more than probable that it is a conventional representation of a divided or halved black calabash or gourd, cut for the purpose of forming it into a cup or dipper, which, in this form, is considered a symbol of this deity.

The evidence upon which this determination is based is that the symbol constantly accompanies the red mouthed, black deity. It is found, with a single exception, only in the Manuscript Troano, and chiefly in Plates II to V, relating to the traveling merchants. The single exception alluded to is on Plate 15 of the Cortesian Codex; here the god bears upon his back the traveling pack, indicating the vocation of which he is the special guardian.

It occurs unconnected with the figure of the deity only on Plates IX*, XIV*, XV*, and XXV* of the Manuscript Troano. In the last the figure of the god is in the same division, but in the adjoining compartment. In Plate XV* it apparently refers to the idol the priest is carving, which is probably a black one intended to represent this god. Landa,[358-1] speaking of the artists carving idols from wood, says:

They took also that which they used for scarifying their ears and drawing blood from them, and also the instruments which they needed for sculpturing their black divinities.

Its appearance in Plate XIV* is apparently in connection with the ceremonies relating to the manufacture of idols. Neither the symbol nor the god it represents is to be fond in the Dresden Codex.



Kukulcan. (?) This is the symbol of the long nosed god, which Dr. Schellhas designates "the god with the snake-like tongue," of which representations appear so frequently in the different codices (see Fig. 381).

The snake-like appendages hanging from the side of the mouth may possibly be intended to represent a curved fang rather than part of a divided tongue. A remarkable figure on Plate 72 of the Borgian Codex deserves special notice here. This is the representation of a deity supposed by Kingsborough and others to be Quetzalcoatl, in which the head is as represented in Fig. 382. Here we see both tongue and fang, and also an eye precisely of the form found in the Maya symbol.



Whether Kukulcan is the god indicated is uncertain, unless he is identical with the long nosed god, or Maya Tlaloc, so frequently figured in the Manuscript Troano and the Cortesian Manuscript. It is only necessary to compare the figures on Plates 2 to 5 of the latter codex with the long nosed, green figures of Plates XXVI, XXVII, XXIX, XXX, and XXXI of the former to be convinced that they represent the same deity, and that this is the Maya Tlaloc or rain god, whatever may be the name by which he was known.

As the symbol which accompanies these is the same as that found in connection with the "snake tongued," long nosed god of the Dresden Codex, there is no doubt that the same deity is referred to. It is worthy of notice in this connection that Plates 29-41 of the Dresden Codex, which are devoted almost exclusively to this deity, refer very largely to water, the god being figured in connection with water no less than twenty-eight times. He is also twice colored black, probably to symbolize the dark rain cloud, and twice blue, denoting water. It is therefore fair to conclude that the author of this codex considered him the giver of rain.



The following reasons given by Dr. Schellhas for supposing that the deity indicated is Kukulcan apparently justify his conclusion, though it is possible some other name may have been applied to him:

He is represented in all the manuscripts, and far more frequently than any other deity. His characteristic marks are always unmistakable. An entire section of the Dresden Codex, pp. 29-43, and pp. 1 and 2, belonging thereto, treat almost exclusively of this god, and wherever he is pictured there we also find his name hieroglyph. He is always characterized by the double, snake-like tongue hanging from his mouth and by the peculiar eye, two marks that are never absent, how numerous and varied soever may be his representations, his symbols, and attributes. We also find him with torches in his hands as symbols of fire; he sits on water; he stands or sits in water or in falling rain; he rides in a boat; he appears in company with a fish as symbol of water or in company of a bird's head as symbol of the atmosphere, upon the day sign Cab as symbol of the earth, sitting, with the ax (machete) in his hand, with arrows or spears, with a scepter, and finally, also, with the body of a snake. Considering the immense variety of this god's representations and the numerous symbols of power in the various elements which the deity rules, we may well be justified in assuming that there are indications here of one of the most important figures in Maya mythology, with one of the principal deities of the people. The most important god of the Mayas was Kukulcan, the creator of the country's civilization, who had come from the far, unknown east, the Mexican Quetzalcohuatl, the Gucumatz of the Kiche, the Kukulcan of the Tzendals. All these names mean "feathered snake," "bird snake." Now, in the above mentioned section of the Dresden manuscript, pp. 29-43, there is found on page 36, middle, the representation of a bird and a snake, the two symbols of the god Kukulcan, which, at the same time, denote his name in the manner of a rebus. That this representation is to be referred to the god with the snake's tongue is rendered probable on the one hand by the fact that this whole section treats of him and is proved on the other hand by the circumstance that in the same place the same snake is found represented with the head of the god; thus, page 35, middle, and 36, above. In the same way this snake with the god's head is also found in the Codex Cortesianus, page 10, middle, a passage which is rendered notable also by the fact that in the writing above the picture there is expressly found as a second sign the name hieroglyph of the god.



Cimi (?). Supposed symbols of the god of death. Occurring very frequently in all the codices, but with several variations (see Figs. 383 and 384).

These are given chiefly on the authority of Drs. Foerstemann and Schellhas, as I have some doubt in reference to this conclusion, for reasons which will here be given.



As Dr. Schellhas remarks, this is "the most characteristic and most easily recognized deity of the Maya Codices"; but this statement will not apply to the symbols, as the variations are such as to render it exceedingly doubtful whether precisely the same idea is embodied in each. Even the two forms here given, both of which are found in all the codices and often together, present variations too marked for us to believe, except upon strong evidence, that they represent the same thing. Nor do the figures of this deity or supposed deity appear to embody throughout the same idea. In fact, they leave us in doubt as to whether any one recognized deity is to be understood. Was there in the Maya pantheon such a deity as the god of death? I have so far been unable to find any satisfactory reason for answering this question in the affirmative.

In the first part of the Dresden Codex, which is devoted, in part at least, if not chiefly, to the maladies of the country, the skeleton figures undoubtedly have reference to death, much like the skull and cross bones in our day. In other places, as Plates XXVII and XXII* of the Manuscript Troano and Plate 7 of the Cortesian Codex, the parched earth appears to be intended, but it must be conceded that here also the idea of death is included. Substantially the same idea, or at least the relation of this god to the earth, appears to be indicated in Plate 8 of the Cortesian Codex, where he is represented as beneath and holding up that upon which another deity, bearing the bread symbol, is seated.

As before stated the two symbols frequently appear in connection, sometimes where the god is figured and often where he is not. It is, therefore, unsafe to conclude as yet that either variety indicates a particular deity known as the god of death.



Symbol of the god with the banded face; seen chiefly in the Manuscript Troano; not found in the Dresden Codex (Fig. 385). This is not the deity which Dr. Schellhas designates as "the god with face crossed by lines."



This deity evidently pertains to the underworld and is closely allied to the so-called god of death. The symbol and the figure are found together in but few instances, yet the peculiar markings are such as to leave no doubt on the mind, that the symbol is intended to denote what is represented by the figure, being simply the head of the deity as invariably figured. They appear together in Plates IIIc, Va, and Vb, XXVIII*c, and XXIXc of the Manuscript Troano, in the first two as having some relation to the traveling merchants, but in the last two in a very different role. The dotted lines with which the bodies of these figures are marked and the peculiar anklets appear to have been introduced to signify relationship to the god of death. Perhaps the most direct evidence of this relation is found in Plate 42 of the Cortesian Codex, where the two deities are brought together at the sacrifice here indicated. The two appear to be united in one in the lower division of Plate XXVI* of the Manuscript Troano.

Figures of this god are also found in some of the Mexican codices, as on Plate 73 of the Borgian manuscript, where the relation to death and to the underworld is too apparent to be mistaken. On Plate 10, same codex, the head of death is marked with the distinguishing black band.

Unfortunately for investigations in this line, the early Spanish notices of the Maya mythology are so brief and confused that we can derive but little aid from them in our efforts to identify the deities figured in these manuscripts. Possibly the one with the banded face may represent Cumahau or Hunhau, the prince of the lower regions; but the role he appears to play where figured, with the exception of Plate II, Manuscript Troano, and Plate 73 of the Borgian Codex, would scarcely justify the name.



(?) Symbol of the deity which Dr. Schellhas designates "the god with the old man's face." Found in all the codices and almost invariably in connection with the representation of the deity shown in our Fig. 386.



The deity denoted by this symbol and by the figure which it accompanies is possibly Zamna or Ytzamna, a deified Maya hero, but the various roles in which he is found make it difficult to decide on this point. He appears comparatively few times in the Dresden Codex, and only in the first few pages. In none of these is there anything to indicate his functions. In Plates 12c and 15c he holds a sun symbol in his hand, which might be supposed to refer to his attributes as "Kinich-Kakmo" but for the fact that the same thing is true of one or two other deities figured in the same codex. In the Manuscript Troano, where he is oftenest represented, his figure and his symbol appear most frequently in connection with the bee or honey industry; for example, on Plate Vc, the only place in the first part of the manuscript where honey appears to be referred to, and twenty-two times in that section of the second part, Plates I* to X*, relating to bees. He also appears to take an active part in the manufacture of idols, engages in painting, aids in the culture or gathering of cacao, engages in predatory excursions, and acts in various other relations. In the left compartment of Plate XXIV*a he bears on his head the head of a bird. In the remarkable double plate (41-42) of the Cortesian Codex he is twice figured, in the central area and at the east (top), and in each case is accompanied by a female deity. In the latter case both god and goddess are bearing in their hands the Kan or corn symbol. In Maya mythology Zamua was given a spouse named Ix Kan-Leox, which signifies the yellow frond or silk of maize.

]

Symbol, according to Dr. Schellhas, of the deity which he names "the god with face crossed by lines," found in all the codices, but most frequently in the Manuscript Troano and the Cortesian manuscript. The deity is usually represented as in Fig. 387.



This is introduced here on the authority of Dr. Schellhas, although I have considerable doubt as to the correctness of his conclusion.

He remarks in regard to it as follows:

Another characteristic and easily recognized deity, which, it is true, is comparatively rare in the Dresden manuscript, but occurs with extraordinary frequency in other codices, and whose sign it is not hard to find, is the god whose face is crossed [surrounded] by peculiar parallel lines, representations of whom are given in the Cortesian Codex (p. 11, below) and Dresden Codex (p. 13, middle). The deity is always male and is found in the Dresden Codex five times, Cortesian Codex eighteen times, Manuscript Troano twenty times, and Codex Peresianus five times.

The sign of this god, as was the case with the others and as seems to be the general rule, consists merely of a representation of the god's head, combined with a sign which probably represents an affix. The sign is found wherever the deity is represented and is an exact rendering of the god's head, so that there can be no doubt as to its being the name hieroglyph. True variations are not found, the hieroglyph being perfectly alike in all the manuscripts.

The nature of this deity is not easily determined, though it occurs in the Codices Troano and Cortesianus with extraordinary frequency, so that it would be seen that these two manuscripts, which evidently belong together, treat principally of this deity. No analogous deity is found in Aztec picture writing. * * * To all appearances we have here a momentous figure of Maya mythology, of which, unfortunately, we know nothing.

It is true that this symbol is found in almost every instance where the figure of the god appears—in fact, with fewer exceptions than others in reference to which there is probably little doubt. It is also true that the symbol is an exact copy of the god's head; but on the other hand there are strong reasons for doubting the correctness of Dr. Schellhas's conclusion.

The first is that the figure of the supposed deity seems to have more indications of being the conventional representation of an idol than of a deity. The lines of the head are precisely the same as those on the heads of the carved idols.[365-1]

We also find it in connection with the wood symbol (marginal No. 6) at the only points where the latter is found in the Cortesian Codex, and, what is significant, in wholly inappropriate places unless connected with an idol figure. These are found in the lower division of Plates 10 and 11, two on the top of thatched roofs and another on the head of the deity called the "god with the old man's face," the head in the latter case being apparently carved from a block of wood.

The second is to the same effect, the symbol being found over each of the figures of the lower division of Plates 26, 27, and 28 of the Cortesian Codex and the middle division of Plates XXXI* and XXXII* of the Manuscript Troano, where there appear to be processions of the different deities. It is also significant that in the latter case each deity is bearing in his hands what seems to be a block of wood from which in all probability an idol is to be carved.

Third, we find rows or lines composed entirely of this symbol, as in the so-called title page of the Manuscript Troano.

DISCUSSION AS TO PHONETIC FEATURES OF THE CHARACTERS.

It must be admitted, as heretofore intimated, that this question has not as yet been satisfactorily answered. Whether what is here presented will suffice to settle this point in the minds of students of American paleography is doubtful; nevertheless, it is believed that it will bring us one step nearer the goal for which we are so earnestly striving. Something is said on this subject in my former work,[365-2] which need not be repeated here.

As it is evident from the preceding list of characters that conventional signs and symbols, often nothing more than abbreviated pictographs, were used in many cases to designate objects and persons, the inference to be drawn, unless other evidence is adduced, is, that this method prevailed throughout. Nevertheless there is some evidence that at the date when these manuscripts were written Maya culture was in a transition state; that is to say, conventional symbols were passing into true ideographs[366-1] and possibly into phonetic characters.

The lack of any satisfactory key to assist us in deciphering them makes it exceedingly difficult to decide how far this change had progressed. We are therefore left wholly to deductions to be drawn from the facts obtained by laborious comparisons of the various relations in which the characters are found and the uses which appear to be made of them in the manuscript.

It will be admitted without question that a large number of these characters are ideographs or conventional symbols, as distinguished from pictures, as, for example, most of those denoting the days, months, and cardinal points. I say most of these, as it is yet possible to learn from some of them the objects they were intended to represent, the characteristic features not being entirely lost, as the symbol for the day Cimi, the "death's head" or skull; that of the day Ymix, "the grain of maize;" that of the month Moan, "the head of the moo or ara," a species of parrot, &c.

It is also possible to show from the manuscripts themselves evidences of the changes from conventional pictographs to true or mnemonic symbols.

Take, for instance, the bird symbols on Plates 16, 17, and 18 of the Dresden Codex, presented in the preceding marginal figures numbered 24, 25, 26, 27, 28, and 33. If the determination be correct as given, it is apparent that, while one of the birds is indicated by the head as a symbol, the others are denoted by ideographs, or by phonetic characters bearing no resemblance to their forms or peculiar features. That numerous examples of this kind are to be found in these manuscripts will be admitted by all who have carefully studied them.

Another fact bearing upon this point is the difference between the Dresden Codex and the Manuscript Troano in regard to marking with symbols the things represented in the pictures. We fail to find in the former (unless that on Plate 30 be a possible exception) the earth or soil represented by any symbol, though frequently occurring in the latter and also occasionally in the Cortesian Codex. The symbol for wood or that appearing so often on wooden articles in the latter, and occasionally in the Cortesian Codex, is wanting in the Dresden Codex, though wooden articles are several times represented. From this we infer that the Manuscript Troano is a more recent production than the Dresden Codex, notwithstanding the evidences of greater skill in drawing and higher mathematical attainments shown in the latter.

Before discussing the question of phonography we ask attention to one or two facts regarding Landa's alphabet which do not appear to have been previously noticed, yet have an important hearing on the subject.

The failure to reach any satisfactory results with this alphabet proves, beyond a reasonable doubt, that this author was mistaken as to the character of the Maya writing; yet the frequent occurrence in the manuscripts of most, if not all, of the elements he presents renders it certain that there is a basis of truth on which it rests. It is probable, therefore, if we can find the key to his method, we may, after all, obtain some satisfactory results by means of his alphabet.

I have already stated as my belief that—

He has undertaken to pick out of their compound or syllabic characters the letter elements; hence it is that, while we find it impossible to decipher the manuscripts by using them, yet we find such frequent resemblances as to compel us to admit a fundamental relationship.[367-1]

This opinion I still believe to be correct, but was, until very recently, unable to get any positive evidence as to his method of obtaining these elements.

While examining the Cortesian Codes I came across (on Plate 17) the symbol for a turtle (the different varieties of which are shown in marginal figure No. 4), which is nothing more or less than an attempt to represent the head of the animal. In the more abbreviated form (b) I at once recognized Landa's A (compare with c and d, No. 4). As the Maya name of the turtle is Ac or Aac it is apparent that in this instance the old Spanish priest selected a symbol representing an object the name of which contains a single syllable having, as its chief letter element, A. As this symbol is simply a representation of the animal's head there is no reason to infer that it is phonetic; on the contrary, it is more reasonable to assume that it was used only as a conventional sign. It is possible that after long usage it may have been adopted as a phonetic character, though its exceedingly rare occurrence in the manuscripts (being found only in the Cortesian Codex and with the turtle figure) and the fact that it is seldom, if ever, used as part of a compound character would seem to forbid this idea.

Precisely the same method was adopted in obtaining his B, which is given in two forms, first as a foot print and second as a circle inclosing four circular dots. The first, as all are aware, is only a conventional sign and presumably not phonetic. The second may be phonetic, though apparently but an abbreviation of the first. In Plate 65c (see marginal No. 20) and Plate 41c the two forms are brought into such relation to each other as to show that the latter is used as a symbol to represent the idea conveyed by the first. The proof in these cases is too strong to admit of doubt and explains Landa's method of obtaining his B, which, as before stated, was by selecting the symbol of that which is denoted by a Maya word of one syllable having B as its chief letter element, Be being the Maya word for "way," "journey," "walking," &c.

The symbol for the cacao given above in marginal No. 22 contains his eleventh letter Ca twice and is probably that from which it was taken; likewise that of the Kukuitz or Quetzal (marginal No. 26) and of the Kuch or vulture (marginal No. 27a), each of which contains his Ku, being double in the former and single in the latter. I am as yet unable to trace these two symbols to their origin; we might suppose, from Landa's figure of the latter, that it was intended to represent a bird's nest containing eggs, but an examination of the symbol as found in the manuscript renders this conclusion doubtful.

The evidences of phonography are few and, as must be admitted, not entirely satisfactory; yet they are apparently sufficient to justify the somewhat general belief that the writing of the Mayas had reached that stage where characters are sometimes used to indicate sounds. That comparatively little advance had been made in this direction at the time of the conquest is possible; moreover there is nothing to justify the belief that they made use of true letters as Landa supposed. If they had a phonographic system of any kind it was very imperfect and was only in that primary stage in which syllables are represented by single characters and words of more than one syllable by compound characters. Judging by the changes observed in the relation of the parts of compound characters to one another, we conclude that the order of arranging these parts was not uniform or essential. It is also doubtful, if any of these characters are phonetic, whether the parts of the longer words were always written out in full. I am led to believe, from a few slight indications, that, in forming words of more than one syllable, they often used only the leading phonetic elements of the single words of which they are composed; in other words, that they followed the rebus method of the Mexicans.

Descending to particulars and examples, the following are, perhaps, the strongest proofs which can be presented on this point:

As there can no longer be any doubt that the symbols for the cardinal points have been ascertained and that those relating to the polar points are distinguishable from those relating to the equatorial points, we are justified in referring to them in this discussion. As each of the two assigned to the equatorial points contains the symbol for "sun" or "day" and as the two Maya words for these points—Likin or Lakin and Chikin—contain the Maya term for sun or day ("kin"), there is some reason for believing that the characters are phonetic. There is to be added to this evidence the fact that the symbol of the month Yaxkin contains the same sun symbol. It would be somewhat remarkable to find the same single character in three different combinations, representing three different ideas expressed by words containing the same sound, yet having no reference to the sound.

It is now generally admitted by students of American paleography, on what appears to be satisfactory evidence, that symbol No. 7 of the preceding list, Cab, is used to signify "earth" or "land" and "honey," both of which are designated by the same Maya term, Cab. As there is no similarity in the things denoted the character is probably phonetic. The "bee" appears also to be frequently indicated by the same character with an affix, as may be seen by reference to the lower divisions of Plates III*—X* of the Manuscript Troano.

The symbol No. 9 (U) of the preceding list is found repeatedly on vases and also as a prefix to both simple and compound characters. As U in Maya signifies "moon," "vase," and certain pronouns and is also used as a euphonic particle before vowels, we are perhaps justified in concluding that the symbol is phonetic and denotes the word U. I am aware that neither Perez nor Dr. Brinton gives "vase" as one of the meanings of this word, yet its constant appearance on vessels seems to leave no doubt that Brasseur is correct. Even admitting that he is mistaken and that we are in error as to the signification of the symbol, its various uses justify the belief that it is phonetic.

The symbol No. 34 of the preceding list, which is supposed to be that of the god Ekchuah, is probably phonetic. The name of this deity is composed of two Maya words, ek, "black," and chu, "calabash," and hence signifies "the black calabash," and the form and coloring of the symbol are apparently intended to denote this signification. If this interpretation be correct it is phonetic, as there is nothing in or pertaining to the figure of the deity which corresponds with it, except the color.

If the interpretation given of the preceding symbols Nos. 22, 24, 26, 27a, and 33 be correct, there can be scarcely a doubt that they are phonetic. In the first—cacau, cacauak, or cacauche, the "cacao"—we see Landa's letter Ca, which is doubled in each of the three forms taken from the different codices. In the twenty-sixth—Kukuitz, the Quetzal—Landa's Ku is duplicated, as it should be if phonetic, while in 27a, Kuch, it appears but once. There is here also an additional evidence of phoneticism in the fact that, while one of the symbols used to denote this bird shows simply its head, and is surely not phonetic, the other is entirely different and bears no resemblance whatever to any feature or characteristic of the bird. Moreover, both parts of it are used in other combinations referring to entirely different things.

If my interpretation of No. 14 (Xamach or Chimix) be right, it is probably phonetic also. It is composed, as will be seen by reference to the figure, of two symbols closely resembling that for the day Ymix, except that the top portion of one is omitted. The resemblance in sound to a duplication of Ymix is apparent. The slight but permanent variation of the right hand portion from the usual Ymix symbol and the omission of the top portion of the left hand one are scarcely explainable on the supposition that they form simply a conventional sign; but if phonetic the reason is apparent, as the m sound is not repeated in the Maya name. This conclusion is strengthened by the fact that the month Mac, found in the last or bottom line of Plate 49, is precisely the same as the right portion of No. 14, with Landa's symbol for Ca added. This probably justifies us in concluding that the true name of this month is Camach, "the jaw" or "jaws," and that Landa's figure is simply a rude representation of the lips or mouth.

I have expressed the opinion[370-1] that the chief phonetic element of No. 8 (the stone symbol), if used to represent sound, is p or pp. This opinion seems to be confirmed by the fact that this character is found as a part of the symbol for the month Pop on Plate 50 of the Dresden Codex. (See the second character in the first transverse line below the day columns in the preceding Fig. 362.) The method of determining the months referred to in these plates of the codex has been given in the preceding part of this paper.

The interpretation given above of symbol No. 24 (the moo or ara) will probably be accepted by all students of these manuscripts, and if so its phonetic character must be conceded. That it is used in the place above alluded to (Dresden Codex, Plate 16c) to denote this bird is proved by the parallelism of the groups and the figure of the parrot under it. If we turn now to Plate 48 of this codex we observe that the second character of the first line below the day columns and the first character in the upper line of the lower group or square is, in each case, a bird's head. It is easily proved by means of the numeral series with which these are connected that they denote, in both cases, the month Moan (from the moo), proving that Brasseur's surmise was correct.[370-2] If the same bird is represented by two symbols, one pictorial and the other having no resemblance to any feature or character of the thing denoted, it is probable the latter is phonetic. This conclusion is strengthened in this case by the strong resemblance of the first part of No. 24 to the symbol for the month Mol.

I have shown above that the right portion of No. 20 of the list is Landa's letter B, and also that in the lower division of Plate 65, Dresden Codex (see Fig. 378), it signifies "footsteps" or the act of walking. As the Maya word Be signifies "journey," "wood," "march," and also "journeying" and "marching," it is possible that this symbol is also phonetic, although apparently only a modified form of the footprint. This supposition is strongly supported by the fact that it is found in numerous and varied relations, single and in combination.

The symbol for 20 (Kal), No. 1 of the preceding list, is apparently phonetic. This view appears to be confirmed by its use otherwise than as a numeral symbol at several points in the text of the Manuscript Troano. For example, in the third division of Plate XVII* it appears in this form, while immediately below is the representation of an idol head in a vessel covered with a screen or basket, as shown in Fig. 388. The Maya verb Kal signifies to "imprison" or "inclose," which is certainly appropriate to what we see in the figure. As the symbol is over each of the three similar figures in the division, it is probable that it is intended to denote something relating to or observable in them. In the second division of Plates XV* and XVI*, same codex, is this symbol, several times repeated, and below each the figure of a priest or deity at work, each carving, with a machete or hatchet, the head of an idol. The probable signification is "Give twice twenty strokes with a machete," and hence is but partially phonetic.



Other examples bearing on this question may be found, but these are believed to be sufficient to warrant the belief that at the time these codices were written Maya culture had reached that stage where the idea of phoneticism was being introduced into the writing. Yet it is certain, and even susceptible of demonstration, that a large portion, perhaps the majority, of the characters are symbols. The more I study these characters the stronger becomes the conviction that they have grown out of a pictographic system similar to that common among the Indians of North America. The first step in advance appears to have been to indicate, by characters, the gesture signs.

FOOTNOTES:

[345-1] See Chapter VI, Study of the Manuscript Troano, by Cyrus Thomas.

[354-1] Unfortunately the scrolls were overlooked in preparing the cut.

[358-1] Relacion de las cosas de Yucatan, p. 308.

[365-1] See Plates XVI*b and XVII*c, Manuscript Troano.

[365-2] Study of the Manuscript Troano, pp. 141-161.

[366-1] As the term "ideograph" is somewhat broad and comprehensive, it may be well enough to state that I use it as expressing that stage of symbolic writing where the picture characters have so changed that all resemblance to the objects they were originally intended to represent is lost, and therefore they can only be considered as mnemonic signs.

[367-1] Study of the Manuscript Troano, by Cyrus Thomas, pp. 142, 143.

[370-1] Study of the Manuscript Troano, p. 147.

[370-2] Landa's Relacion, pp. 382, 383, Note 1.



INDEX

Adelung, J. C. cited 262

Aglio, Augustina, fac simile of Dresden Codex by 263-266

Boettiger, C. A., mention of Dresden Codex by 262 controversy with Abert concerning Dresden Codex 267

Brasseur, copy of the Manuscript Troano by 284, 286, 343 cited 350

Calendar system, tabular view of 270-374

Charency, H. de, cited 282

Codex Cortesianus, similarity of, to Manuscript Troano and Dresden Codex 286

Dresden Codex, numerals in 261-338

Ebert, F. A., description of Dresden Codex by 263 controversy with Boettiger concerning Dresden Codex 267

Falkenstein, K. C., preservation of Dresden Codex by 268

Fleischer, H. L., mention of Dresden Codex by 263

Foerstemann, E., citation from Die Mayahandschrift of 261-269 cited 272, 278, 280, 281, 283, 290, 292, 293, 300, 301, 302, 303, 304, 305, 320, 322, 329, 330, 339, 340

Goetze, J. C., preservation of Dresden Codex by 261 biographical sketch of 261, 262

Humboldt, A. von, notice of Dresden Codex by 262, 263

Kingsborough, Lord, Dresden Codex copied by order of 262 Mexican Antiquities of, cited 266

Landa, cited 348

Landa's alphabet, insufficiency of 259, 347

Manuscript Troano, copy of, by Brasseur 285, 286, 343 study of, by C. Thomas, cited 339, 343, 344, 345, 350, 365, 366, 367, 370

Maya and Mexican manuscripts, C. Thomas on, cited 280

Maya Codices, aids to the study of, by C. Thomas 253-371

Mexican Antiquities, by Lord Kingsborough, cited 266, 267

Rosny, L. de, cited 267, 347, 355, 357

Schellhas, P., cited 345, 359, 360, 361, 362, 364

Schultz-Sellack, K., cited 278

Silvestre, E., Paleographie universelle of, cited 267

Thomas, C., paper on aids to the study of the Maya codices by 253-371

Troano Manuscript, copy of, by Brasseur 285, 286, 343

Vater, J. S., cited 262



Transcriber's Note:

TN-1 267 "hasty and obtrusive notice." should read 'hasty and obtrusive notice.' TN-2 272 indi cated should read indicated TN-3 291 "and" repeated. TN-4 295 Plate 48, 2nd line first asterisk is missing. TN-5 296 Period missing after FIG TN-6 322 In the original text, the 7 is printed above the 17, with no horizontal line separating the two numbers. TN-7 327 Tables XXI and XXII are not labeled in the original publication but, by context, appear to be the two sections of table following Table XX. TN-8 338 Sixth column should read Sixth column. TN-9 338 Footnote number for 338-1 was missing. It was inserted based on the context of the note. TN-10 348 Illustration No. 6 was missing the caption. TN-11 364 Fig. 39 should read No. 39

THE END