SUBROUTINE MHIST(MM, M, N, A, CLAB, RLAB, TITLE, XMIN, XMAX, NV,
     *                 NDIAGS, DMWORK, WORK, DMIWRK, IWORK, IWORK2, XLL,
     *                 OUNIT)
C
C<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
C
C   PURPOSE
C   -------
C
C      PRODUCES MULTIVARIATE HISTOGRAMS
C
C   DESCRIPTION
C   -----------
C
C   1.  THE DIAGRAMS ARE COMPUTED SEQUENTIALLY.  THE FIRST DIAGRAM
C       BEGINS WITH THE SMALLEST RANGE FOR EACH VARIABLE WHICH CONTAINS
C       EVERY OBSERVATION.  THEN THE RANGES FOR EACH VARIABLE ARE
C       VARIED AND THE NEW RANGES ARE ACCEPTED IF THEY INCREASE THE
C       LOG-LIKELIHOOD OF THE OBSERVATIONS WITHIN THOSE RANGES.  EACH
C       RANGE IS VARIED UNTIL NO FURTHER CHANGE INCREASES THE
C       LIKELIHOOD.  THE FIRST DIAGRAM IS COMPLETED AND PRINTED AND THE
C       OBSERVATIONS WITHIN THE FINAL RANGES ARE DELETED AND THE VOLUME
C       OF THE BLOCKS ARE IGNORED IN LATER DIAGRAM CALCULATIONS.  A
C       SECOND DIAGRAM, AGAIN STARTING WITH THE SMALLEST RANGE FOR EACH
C       VARIABLE WHICH CONTAINS EVERY REMAINING OBSERVATION, IS ADDED
C       AND VARIED TO MAXIMIZE THE LIKELIHOOD OF BOTH BLOCKS.  THE
C       OTHER DIAGRAMS ARE COMPUTED SIMILARLY.
C
C   2.  THE OUTPUT IS ON FORTRAN UNIT OUNIT AND CONSISTS OF THE SET OF
C       DIAGRAMS.  FOR EACH DIAGRAM, THE RANGES FOR EACH VARIABLE ARE
C       ENCLOSED IN A RECTANGULAR BLOCK.  TO THE RIGHT OF EACH DIAGRAM
C       IS A RECTANGLE WHOSE LENGTH IN THE HORIZONTAL DIRECTION IS THE
C       NUMBER OF CASES WITHIN THE SET OF RANGES FOR THAT PARTICULAR
C       DIAGRAM.  THE SUM OF THE HORIZONTAL DISTANCES FOR ALL DIAGRAMS
C       SHOULD ADD UP TO THE TOTAL NUMBER OF CASES.
C
C   INPUT PARAMETERS
C   ----------------
C
C   MM    INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         THE FIRST DIMENSION OF THE MATRIX A.  MUST BE AT LEAST M.
C
C   M     INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         THE NUMBER OF CASES.
C
C   N     INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         THE NUMBER OF VARIABLES.
C
C   A     REAL MATRIX WHOSE FIRST DIMENSION MUST BE MM AND WHOSE SECOND
C           DIMENSION MUST BE AT LEAST N (UNCHANGED ON OUTPUT).
C         THE MATRIX OF DATA VALUES.
C
C         A(I,J) IS THE VALUE FOR THE J-TH VARIABLE FOR THE I-TH CASE.
C
C   CLAB  VECTOR OF 4-CHARACTER VARIABLES DIMENSIONED AT LEAST N.
C            (UNCHANGED ON OUTPUT).
C         THE LABELS OF THE VARIABLES.
C
C   RLAB  VECTOR OF 4-CHARACTER VARIABLES DIMENSIONED AT LEAST M.
C            (UNCHANGED ON OUTPUT).
C         THE LABELS OF THE CASES.
C
C   TITLE 10-CHARACTER VARIABLE (UNCHANGED ON OUTPUT).
C         TITLE OF THE DATA SET.
C
C   XMIN  INTEGER VECTOR DIMENSIONED AT LEAST N (UNCHANGED ON OUTPUT).
C         XMIN(I) IS THE MINIMUM VALUE TO BE PLOTTED FOR VARIABLE I.
C
C   XMAX  INTEGER VECTOR DIMENSIONED AT LEAST N (UNCHANGED ON OUTPUT).
C         XMAX(I) IS THE MAXIMUM VALUE TO BE PLOTTED FOR VARIABLE I.
C
C         IF XMIN(I) .GE. XMAX(I) ON INPUT, THEIR VALUES WILL BE
C            DETERMINED BY THE ROUTINE.
C
C   NV    INTEGER VECTOR DIMENSIONED AT LEAST N (UNCHANGED ON OUTPUT).
C         VECTOR DEFINING THE ORDER OF THE VARIABLES.
C
C   NDIAGS INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         THE NUMBER OF DIAGRAMS.
C
C   DMWORK INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         THE FIRST DIMENSION OF THE MATRIX WORK.  MUST BE AT LEAST
C            2*NDIAGS.
C
C   WORK  REAL MATRIX WHOSE FIRST DIMENSION MUST BE DMWORK AND WHOSE
C            SECOND DIMENSION MUST BE AT LEAST N+1.
C         WORK MATRIX.
C
C   DMIWRK INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         THE FIRST DIMENSION OF THE MATRIX IWORK.  MUST BE AT LEAST
C            2*NDIAGS.
C
C   IWORK INTEGER MATRIX WHOSE FIRST DIMENSION MUST BE DMIWRK AND WHOSE
C            SECOND DIMENSION MUST BE AT LEAST N.
C         WORK MATRIX.
C
C   IWORK2 INTEGER VECTOR DIMENSIONED AT LEAST M.
C         WORK VECTOR.
C
C   OUNIT INTEGER SCALAR (UNCHANGED ON OUTPUT).
C         UNIT NUMBER FOR OUTPUT.
C
C   OUTPUT PARAMETER
C   ----------------
C
C   XLL   REAL SCALAR (CHANGED ON OUTPUT).
C         THE LOG LIKELIHOOD OF THE FINAL SET OF BLOCKS.
C
C   REFERENCES
C   ----------
C
C     HARTIGAN, J. A. (1975).  CLUSTERING ALGORITHMS, JOHN WILEY &
C        SONS, INC., NEW YORK.  PAGES 40, 54-55.
C
C     HARTIGAN, J. A. (1975) PRINTER GRAPHICS FOR CLUSTERING. JOURNAL OF
C        STATISTICAL COMPUTATION AND SIMULATION. VOLUME 4,PAGES 187-213.
C
C<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
C