GrassBase - The Online World Grass Flora


W.D. Clayton, M.S. Vorontsova, K.T. Harman & H. Williamson

© Copyright The Board of Trustees, Royal Botanic Gardens, Kew.


The interactive program is accessed through an icon; terminated by QUIT. It uses binary data files, YCHARS and YITEMS, prepared by CONFOR. The parameter file INTKEY.INI is read automatically on entry. A taxon list can be obtained from YNAMES.PRT.

At the first dialogue select the required data set (or enter the path to an .INI file). The initial screen is set up for automatic identification only. For manual operation, with all functions available, click File>Advanced Mode.

Commands can be entered in three ways:

  1. At the menu bar.

  2. On the toolbar.

  3. On the command line (abbreviations permitted).

Pressing ESC will abort most commands by successively closing windows;some commands provide a STOP button instead.

Ranges and Keywords

Most commands require a range of taxa or characters to be specified. This can be done singly, or in groups designated by the following keywords.

Character keywords

ALL NONE USED AVAILABLE and predefined character groups. Characters dependent on an unused character are not treated as available.

Taxon keywords

ALL NONE REMAINING ELIMINATED SPECIMEN (in process of identification) and "taxon names". Further groups may be user-defined, but are not pre-set.


The command opens a keyword list; then OK to enter all the keyword taxa/characters, or LIST followed by deselect-all, to make a selection. The selection can be extended by shuttling back to the keyword list.

Command line

Entry takes the form of (taxon numbers); character numbers; character,state combinations; or "taxon names".Parentheses and quotes can be omitted if no blanks are included. The conjunctions -(to) and /(or) can be used. Text characters have "string" values, with leading or trailing blanks if desired;they are case insensitive.


Identification and interrogation proceed by eliminating taxa lacking the required attributes. Three strategies can be pursued:

  1. Enter obvious or unusual characters to reduce size of data set.

    1. USE opens a keyword menu. Command line entry for example character or state can omit use by being directly entered onto the command line.

    2. CHANGE/DELETE allows characters in use to be changed or deleted. They can also be changed by re-entry on the command line.

  2. Invite program help.

    1. BEST suggests best characters for dividing remaining taxa, their efficiency being assessed by a separation value.

    2. SEPARATE(taxon number) suggests quickest characters when identity can be guessed.

    3. LASTBEST redisplays last BEST/SEPARATE list.

  3. Make critical comparisons. See description menu below.

    RESTART readies program for next identification.

  1. DESCRIBE lists character states of designated taxa.

  2. DIFFERENCES lists differences between designated taxa.

  3. SIMILARITIES lists similarities between designated taxa.

  4. DIAGNOSE provides each of the designated taxa with a character set which differs in at least DIAGLEVEL characters from all other INCLUDEd taxa(restrict these or it goes on for ever). Pre set characters will appear in every diagnosis; their presence speeds up the operation (default none).

  5. MATCH is temporarily set to OUI.

  6. SUMMARY reports the distribution of character states in a given char/taxon set.

  1. CHARACTERS/TAXA opens a keyword window, then either

    1. OK for a full window display of keyword characters and their states; or

    2. LIST for a compact character list. Only AVAILABLE characters are displayed, unless USED is a keyword.

      Taxon lists show number of differences from specimen in parentheses.

    3. CHARACTERS/TAXA <range> on command line is quicker.

  2. FIND CHAR/TAXA <string>Case insensitive.

  1. INCLUDE/EXCLUDE CHAR/TAXA specifies the data set on which subsequent commands operate. It overwrites any previous INCLUDE/EXCLUDE command.

  2. DEFINE TAXA <"keyword"range/previously defined keywordsrds> Gives a keyword name to a group of taxa.


    1. U unknowns match any value.

    2. I inapplicables match any value (unless a difference is implied by preset dependency declarations).

    3. O values match when ranges overlap.

    4. S values match when one range is a subset of (included in) the other.

    5. E values must match exactly.

      The default setting OUI is usually best. Use MATCH O for SIMILARITIES;and for information retrieval (otherwise unknown states will count as hits).

  4. SET FIX ON /OFF fixes a partial identification;

    1. RESTART returns to this point for further identifications.

  5. SET TOLERANCE n - the number of differences permitted before a species is eliminated. Default 0.

  6. SET DIAGLEVEL n - the number of diagnostic differences to be sought for each taxon. Default 1.

  7. SET EXACT. Characters regarded as not subject to error when TOLERANCE > 0.

  8. SET RELIABILITIES char,r - where r = 0-10. Defaults are pre set in INTKEY.INI

  9. SET STOPBEST n - where n = number of characters to be examined by BEST (default 50).It starts with the most reliable characters, and sorts by descending separation value. Press Enter (without selection) to extend the search by another n characters.

  10. SET RBASE n - where n = 1(ignored) - 5(paramount)governs the effect of reliabilities when searching for BEST.

  11. SET VARYWT n - where n = 0(variable characters ranked low) - 1(no penalty)governs the treatment of characters which vary within species; essentially whether BEST may split variable species or not.

  12. STATUS reports current parameter settings.


  1. FILE OUTPUT<filename> output is echoed to an append file. For long lists it is best to switch off screen pagination (DISPLAY CONTINUOUS ON/OFF). OUTPUT commands always go to a file, which must have been declared but need not be on; turning it off (OUTPUT OFF/ON) avoids collecting unintended clutter from screen display.

  2. OUTPUT COMMENT<text> - text is copied to the output file; used to insert headings or directives.

  3. OUTPUT CHAR/TAXA outputs a list of char/taxon numbers, suitable for use as a CONFOR parameter.

  4. OUTPUT DESCRIBE outputs a Delta format description.

  5. OUTPUT SUMMARY outputs a combined Delta format description for a designated group of species, such as a genus. Each state has a comment of the form <n/m>, where n = number of species exhibiting the state and m = number for which character recorded. The comment is reduced to <n> when all taxa recorded;and omitted when all species have the same state.

  6. OUTPUT DIAGNOSE outputs diagnostic sets as a list of char/taxon numbers; suitable for CONFOR directive EMPHASIZE.

  7. FILE DISPLAY<filename> displays file in a separate window.

  8. FILE CLOSE<filename> releases a file from the program; it may be reopened. Closure is automatic on QUIT.


Accesses a log file of previous commands, with facility for repeating them.


Window management. Cascades or tiles or closes open Windows.