Library of Congress

Program for Cooperative Cataloging

The Library of Congress > Cataloging, Acquisitions > PCC > CONSER > Publication Patterns Initiative > Instructions for Record Captions and Pattern Data

Structure of the field

First indicator -Compressibility and expandability

0 Cannot compress or expand
1 Can compress but not expand
2 Can compress or expand
3 Unknown

Second indicator--Caption evaluation

0 Captions verified; all levels present
1 Captions verified; all levels may not be present
2 Captions unverified; all levels present
3 Captions unverified; all levels may not be present

Subfields

a First level of enumeration (M)
b Second level of enumeration (NR)
c Third level of enumeration (NR)
d Fourth level of enumeration (NR)
e Fifth level of enumeration (NR)
f Sixth level of enumeration (NR)
g Alternative numbering scheme, first level of enumeration (NR)
h Alternative numbering scheme, second level of enumeration (NR)
i First level of chronology (RA)
j Second level of chronology (NR)
k Third level of chronology (NR)
l Fourth level of chronology (NR)
m Alternative numbering scheme, chronology (NR)
n Pattern note (NR)
o Type of supplement or index
pNumber of pieces per issuance (NR)
u Bibliographic units per next higher level (RA)
v Numbering continuity (RA)
w Frequency (NR)
x Calendar change (NR)
y Regularity pattern (RA)
z Numbering scheme (R)
3 Materials specified (NR)
8 Link and sequence number (NR)

Note: Subfield 6, Linkage, will not be used in the project.

Back to Top

5.1. Indicators

Indicators are used only with basic volumes and supplements. For indexes, the indicators are blank.

The first indicator tells an automated system whether data may be compressed (expressed as a range beginning with the first enumeration/chronology and ending with the last enumeration/chronology) or expanded (expressed in itemized format). The ability to compress and expand where lower levels of caption are present depends on the presence of needed pattern subfields $p-y, while a holdings record with a single level of caption may need only the frequency code $w.

The second indicator indicates whether or not the captions are verified, and whether all levels of caption have been included.

For indexes, the indicators are blank. The enumeration subfield of an index field (891 $9 855) is the enumeration of the bibliographic units covered. These cannot be compressed without making an ambiguous string.

For 891 $9 853 and 891 $9 854, use the indicators specified above.

CONSER note: Many systems preformat MARC holdings data with default indicators. It is not mandatory that libraries try to input correct indicators into the CONSER 891 fields. However, in instances where a correct publication pattern is being input from verified serials control or issue-in-hand information, an inputting library may wish to signal the completeness and the correctness of the data. In these cases, the indicators would be 20. Without these guarantees, indicators might be 33:

891 20 $9 853 [must have all appropriate captions, $a-m, and all appropriate pattern subfields $p-y that may be needed to govern prediction.]

891 33 $9 853 [compressibility and expandability are unknown, and the inputter has not verified the captions against the actual pieces.]

5.2. Subfields.

5.2.1. Subfield $9 -Tag

Subfield $9 is input as the first subfield of field 891. Its content is the tag of the destination MARC holdings field. In the captions/pattern field, the values are:

$9 853 captions/patterns for base volumes
$9 854 captions/patterns for supplements
$9 855 captions/patterns for indexes

5.2.2. Subfield $8 - Link and sequence number

Subfield $8 is input as the second subfield of field 891. In the caption/pattern field, it contains a linking number which will determine its sequencing for output. For fields with identical subfields $9, the first will be $8 1, the second $8 2, the third $8 3, etc. So that the number of pattern changes can be monitored, when inputting a new caption and pattern field (891 $9 85X) to a record where a previous pattern was recorded, assign the next sequential linking number for the specific destination tag, even when older patterns will be deleted. See Section 5.3, "When to input a new caption/pattern field."

Example

Back to Top

Defining and inputting the enumeration captions

5.2.3. Subfields $a-h - Enumeration

In the two 891 fields, subfields a-h carry the various levels of enumeration of the bibliographic units of a serial. The first field of the pair contains the captions for that enumeration.

5.2.3.1. Examine the sources

To determine the caption information for a serial, examine within the latest issue the sources of bibliographic information that would be examined in order to determine the designation (CCM, Module 8.3). Refer to other issues (previously checked in) to help identify the relationship of the enumeration and/or chronology to the parts published. Use designation information in the CONSER bibliographic record as a check on your findings and keep in mind that it also may need to be updated. Pay particular attention to the following elements of enumeration and chronology:

5.2.3.2. Identify primary bibliographic units

Identify the highest level of enumeration and/or chronology, often "volumes" or an equivalent term. Note the name of the unit and its interval of publication. Note whether it has further internal units. If a serial is issued with multiple levels of enumeration, the higher unit usually increments on an annual basis. However, some serials have 2 or more volumes per year; others have a new volume every few years. There may or may not be dates associated with one or with each level of part.

Back to Top

5.2.3.3. Identify any secondary or lower bibliographic units

Consider whether they represent lower levels or alternative numbering schemes, which are separate sequences from the primary hierarchy of enumeration.

In serials with multiple levels of enumeration, internal part numbering will usually "restart" when the primary bibliographic unit changes (increments). In some cases, however, both numbers will increment continuously:

v. 1, no. 1   v. 1, no. 2   v. 1, no. 3   v. 1, no. 4   v. 2, no. 5

Enumeration commonly has captions preceding the number. The following exceptions are common:

A bibliographic unit may bear numbering alone, with no captions.
Examples:

1 2 3     1-1 1-2 1-3

Note: If there are two captionless numbers combined by a symbol as in the second grouping, it could be a single consecutive numbering, or a more inclusive numbering followed by internal divisions. If the former, the internal hyphen will be changed to a period, but otherwise the numbering is used as given. If the latter, it should be divided into two subfields, both without caption. See Section 6.2.3., "Defining and inputting the enumeration and chronology, Subfields $a-m Enumeration and chronology data."

A bibliographic unit may bear an ordinal number which precedes the caption: 1st ed.

A bibliographic unit may bear a date, whether of coverage or issuance. Secondary and lower bibliographic units can be chronological (e.g., month, month/day, season) or numeric.

Examples:
A serial bears a designation on the cover consisting of the current month and year. This data is input in the caption subfields as $a (year) $b (month)

A serial bears a designation consisting of the year with internal numeric divisions (e.g., no. 1, 2, 3, 1998; no. 1, 2, 3, 1999). This data is input in the caption subfields as $a (year) $b no. (or qtr., issue, trimestre, etc)

Back to Top

5.2.3.4. Input the captions for each level of enumeration in subfields $a through $f

Keep in mind that each lower caption will be followed by its pattern. ($u-v).

Special considerations:

If the designation of the bibliographic unit is an ordinal number which precedes the caption, place a + before the caption as in the third example. Note: do not use the + sign merely to place the enumeration in advance of the caption; use it only when denoting an ordinal number. If the statement is, for example, 1998 ed., either give ed. without the + sign, in which case the syntax will be changed to "ed. 1998", or drop the specific caption and give the date with the usual caption (year).

If the serial bears enumeration that lacks captions, supply in each such subfield an asterisk in parentheses (*) to serve as a placeholder, as in the fourth example.

If the serial is designated only with chronological designations, input these also in subfields $a-f, enclosed in parentheses to suppress their display, as in the fifth example.

Note that a year, and in some cases a non-calendar (e.g., academic) year may also have internal numbering as in the sixth example.

If a level of enumeration is alphabetic rather than numeric, it may be input as found, using subfield $z to define the scheme, as in the seventh example. See Section 5.2.11.

If the highest level of enumeration contains a numbered or unnumbered series designation (i.e. Series 2, or New Series), input the series designation together with the following abbreviation, separated by a colon, as in the eighth example.

Use the standard abbreviations in Appendix A. Use v. for volume.

Examples

Optionally, caption data for alternative numbering, if found, is included in subfields $g and $h. If data is input, it is placed after the primary and secondary enumeration captions or the last enumeration pattern subfields ($u-v) (see next section 5.2.3.5.) Remember that it may be omitted if not considered useful, and that it is limited to two subfields.

891 20 $9 853 $8 1 $a v. $b no. [pattern] $g session $h section $i (year)

5.2.3.5. Defining and inputting the publication pattern

CONSER note: Publication pattern information for lower levels of enumeration/chronology are input in subfields $u, $v. Three other pattern subfields, $w, $x, and $y, hold pattern values that pertain to the whole publication (its frequency, its point or points of calendar change, and its regularity pattern).

If pertinent, an additional pattern subfield, $p, number of pieces per issuance, may also be added. See Section 5.2.7 (Number of pieces per issuance).

Back to Top

5.2.4. Subfield $u -Bibliographic units per next higher level

Identify for any secondary and lower levels of enumeration the number of parts issued before the number of the level directly above increments. Input this number in subfield $u.

For example, if there are twelve issues in Volumes 1, 2, and 3, the number of secondary units per primary unit is 12, and subfield $u is 12. When 12 units are published in volume 4, that number will increment, and the next issue to appear should (granting no change in pattern) bear the volume number 5 as part of its designation. Each level of bibliographic unit has its own $u, which may have the value var or und if the value varies or is unknown.

Examples

5.2.5. Subfield $v - Numbering continuity

Determine for any secondary and lower levels of enumeration their behavior when the level above increments. Input this number in subfield $v. For example, when Volume 1 increments in the first example in Section 5.2.3.3., the secondary level of enumeration restarts if Vol. 2, no. 1 is the next issue; it continues if the next issue is Vol. 2, no. 5.

Examples

5.2.6. Subfields $ i-m - Chronology captions

Input chronology relating to the enumeration as shown on the piece, if any, in subfields $i-m. Because chronology captions are not displayed, they should be given in parentheses. As with enumeration, the largest, most inclusive unit of chronology (normally a year) is given first. In some cases, the chronology subfields will correlate to the enumeration subfields. This is not required.

Examples

N.b. Subfield $m is reserved for the chronology of the alternative numbering, if any.

5.2.7. Subfield $p - Number of pieces per issuance

If an issue regularly consists of a number of parts or pieces differing from the publishing frequency expressed in subfield $w (for example, a semiannual publication with six numbers published at each time), a subfield $p, specifying the number of pieces per issuance, can also be added to the pattern data.

Examples

5.2.8. Subfield $w - Frequency

Determine, as closely as possible, the frequency (publication interval) of the serial. Input this value in subfield $w. This subfield is input after the last chronological caption used.

a annual
b bimonthly
c semiweekly
d daily
e biweekly
f semiannual
g biennial
h triennial
i three times a week
j three times a month
k continuously updated
m monthly
q quarterly
s semimonthly
t three times a year
w weekly
x completely irregular

If there is no applicable frequency, note whether there is a standard number of issues per year.

Examples

5.2.9. Subfield $x - Calendar change

Determine, for the entire serial, at what point or points within the year the numbering of the primary bibliographic unit increments. Input this value in subfield $x using the codes as given below. Separate any multiple codes by commas; any values from 1-9 are left-justified with a zero in the unused position.

Possible values
(months) 01-12
(days) 01-31 (combined months/days: 0101 - Jan. 1)
(seasons) 21 (spring), 22 (summer), 23 (fall), 24 (winter)

Special procedures for seasonal chronology: If the year begins with any season other than spring (21), use the 853 $y to indicate the order of seasons within the year. See the last example in Section 5.2.10, Examples.

Examples

Back to Top

5.2.10. Subfield $y - Regularity pattern:

A regularity pattern subfield is needed only for serials whose periodicity varies from the normal pattern for the frequency by regular omissions or combined numbers. The subfield may be structured in more than one way, using chronology or enumeration code values (or both). This subfield is composed of codes which are in three parts:

a publication code:

c combined
p published
o omitted

a chronology code definition:

d day
m month
s season
w week

a chronology code:

Table I. Chronology Codes

Chronology Code Definitions Pattern Chronology Code Meaning
d dd su Sunday
m MM 09 Sept.
d MMDD 0925 Sept. 25
s SS 22 Summer
w wwdd 03we 3rd Wed. of month
w MMWWdd 0599tu last Tue. in May
w MMwwdd 0802we 2nd Wed. in Aug.

Table 2. Chronology Pattern Definitions

Pattern Code Type Alpha/numeric Possible values
dd day alpha mo-su
DD day numeric 01-31
ww week numeric 01-53
WW week numeric 99, 98, 97, 00, 01, 02, 03, 04, 05
MM week numeric 01-12
SS season numeric 21, 22, 23, 24

Back to Top

Table 3. Specific Value Definitions for Chronology

Day of Week Week Month
mo - Monday 01-53 01-12
tu - Tuesday    
we - Wednesday    
th - Thursday    
fr - Friday    
sa - Saturday    
su - Sunday    
  Week of Month Season
  99 - Last 21 - Spring
  98 - Next to Last 22 - Summer
  97 - Third to Last 23 - Autumn
  99 - Last 21 - Spring
  00 - Every 24 - Winter
  01 - First  
  02 - Second  
  03 - Third  
  04 - Fourth  
  05 - Fifth  
Day of Month    
01-31    

Conventions-Chronology:
The chronology type code is structured as: <Publication Code><Chronology Code Definition><Chronology Code>,< Chronology Code>
Two-character alphabetic or numeric codes are used for days, weeks, months, and/or seasons. A code of less than two digits is right justified and the unused position contains a zero. Multiple codes are separated by a comma. A slash (/) is used to designate combined issues. Values can be repeated, if multiple issues are to be received within the same defined time period. Chronology codes are input in the order in which they occur wthin the calendar year.

Conventions-Enumeration:
The enumeration type code is structured as <Publication Code><Enumeration Code Definition><Enumeration Code>,< Enumeration Code>
The enumeration code definition indicates whether the subsequent codes represent enumeration rather than chronology. When code "e" is used, an additional numeric code is added to indicate the level of enumeration to which the regularity applies:

e1 enumeration, first level
e2 enumeration, second level

The enumeration code indicates the designation of the issues of the item for which regularity pattern information is provided. Multiple designations are separated by a comma. A slash (/) is used to designate a combined issue.

Note that enumerative values for titles which have continuous numbering (subfield $v, code c) must be entered in terms of the number of issues expected to be published to allow issue prediction to occur.

The Regularity pattern codes are controlled by an authoritative list maintained at the Library of Congress. Guidelines for their use can be found at: http://www.loc.gov/marc/chrono_patterns.html. Questions on coding patterns or establishing new ones may be addressed by contacting the Network Development and MARC Standards Office (ndmso@loc.gov)) or the CONSER Pattern Holdings Task Force (CONSERHOLD-L@cornell.edu).

Examples

Back to Top

5.2.11.Subfield $z. Numbering scheme

Subfield $z contains a six-character code string used to designate the numbering scheme on a publication. The codes allow for recording different numbering schemes at different levels of enumeration. There are three elements to the numbering scheme: type of designation, case, and script code/type code.

The first one-character code indicates whether the number is a number, letter, or combined (number first, or letter first). Combined should be used only when one of the elements is a common designation (e.g., 1a, 2a, 3a), rather than two different levels of enumeration (e.g., 1a, 1b, 1c).

Designation types:
a Number
b Letter
c Combined, number first
d Combined, letter first
e Symbol or special character

The second one character code indicates if a numbering scheme is conveyed as alphas and applies both to those coded in the previous position as "b" or to Roman numerals.

a No case
b Lower case
c Upper case
d Mixed case

The third through sixth positions indicate with the script used in the numbering scheme or, for numerals or symbols that are not in alternate scripts, the type of numeral or symbol used. The script code is a four character code from the list at: http://www.unicode.org/iso15924/.

an Arabic numeral
rn Roman numeral
sy<symbol> Symbol

Examples

Back to Top

5.2.12. Subfield $3 Materials specified

Optionally: Subfield $3 will be used to show the latest issue consulted if it is necessary to state explicitly that a pattern is still valid, e.g., after minor changes.
CONSER note: Although in most of the MARC formats, subfield $3 generally appears at the beginning of the field, its placement in field $891 will be at the end of the field.

Examples

Subfield $3 will also be used to explain the nature of variations which do not require the input of a new caption and pattern field. See Section 5.3 following.

5.2.13 Subfield $n Pattern note

Optionally, subfield n can be added as an explanatory note citing the specific year and issue used to prepare the regularity pattern coding in the field.

Examples

5.3. When to input a new captions/pattern field

5.3.1. Changes in captions.

With a few exceptions, each change in captioning requires the input of a new 891 $9 85X with a new linking number. For example, if the captions change from Band x, Heft x to Volume x, No. x, a new 891 $9 853 is input.

In record: 891 $9 853 $8 1 $a Bd. $b Heft ...

New field: 891 $9 853 $8 2 $a v. $b no. ...

Exception : If the change is only the addition of a lower level of caption, the new caption may be simply added to the field with any appropriate pattern information, if this can be done intelligibly. Judgment must be used in deciding whether to include data about this change in subfield $3:

891 $9 853 $8 2 $a v. $b no. [pattern] $c pt. [pattern] $3 v.6- ; added pts., v. 7-

Back to Top

5.3.2. Changes in pattern

Changes in publication pattern of the issues also require input of a new 891 field.
Often, when one element of pattern changes, other elements change as well.

Changes in the frequency of the serial ($w)
Changes in the number of parts per volume ($u)
Changes in the numbering continuity ($v)

5.3.3. Noting minor variations

Some variations in pattern may be able to be adequately covered by notes given in subfield $3. This practice will free up more of the record for cataloging and holdings information, and will limit the amount of data which has to be removed from the record if it grows to exceed the size limitations. Also, some changes which occur may not represent true intention of the publisher to change the way the serial is issued; some may even turn out to be misnumberings. Changes may be ignored, if minor; if it is necessary to explain them, the explanation can be given in subfield $3.
As a rule, regular combined issues are coded in the publication pattern. BUT, if they appear irregularly, combined issues are ignored; irregularity may be noted in subfield 3, e.g. (Some issues combined).

If an existing serial begins to be published with one or more regular combined issues, a change may be made in the existing publication pattern to show the new scheme, but the inputter may decide not to input a new field until the change is certain. Subfield $3, in addition to giving the latest issue consulted, explains the variation in the pattern and specifies the point when the change occurred. Here is a quarterly serial with combined issues Spring/Summer, Fall/Winter, where the first volume in hand had four separate issues:

Example

As a rule, the calendar change (point in the year when the enumeration of the base volumes increments) is coded in the publication pattern. BUT, if no pattern of calendar change can be perceived, the inputter does not input a value in subfield x.

If the only change in the pattern is the calendar change point, the inputter changes the value in the calendar change subfield to the new value, but does not input a new field. The old value may be noted in subfield $3.

Example

Do not update a record for minor changes, such as merely to record the last issue consulted, though updating this value is helpful if the record is being worked on for other reasons.

Back to Top

5.4. Inputting caption and pattern data for supplements (Optional)

Supplements, if published on a regular schedule and considered particularly important for the publication pattern, may have data coded in 891 fields. The procedures given above should be applicable to supplements. If supplements do not have their own numbering, because they are supplements to base volumes, their captions may be given as such:

891 20 $9 854 $8 1 $a Suppl. to v. $i (year) $w a $3 suppl. to v.9 LIC
(for a supplement to a serial volume appearing annually)

A supplement that needs a further description may have a subfield $o (Type of supplement) giving the descriptive wording:

891 20 $9 854 $8 1 $a (year) $o buyer's guide $w a $3 1998 LIC

Be careful to choose a new link and sequence number ($8 ) when inputting an 891 field on a record where an 891 field with the same destination tag ($9) already exists.

5.5. Inputting caption and pattern data for indexes (Optional)

5.5.1. Cumulative indexes, if published on a regular schedule and considered particularly important for the publication pattern, may have data coded in 891 fields. There is one difference from the procedure for coding basic volumes and supplements, in that no indicators are used. Many cumulative indexes do not have their own numbering, but instead bear the numbering of the volumes covered by their indexing. If desired, this fact can be made clear by giving the caption as "Index to v.:" (or equivalent) As with supplements, the type of index can be specified in subfield $o. In subfield $3, a slash is used between the designations of the first and last volume or date covered by the index.

Examples

Questions? Contact the Project at: CONSERHOLD-L@cornell.edu

Back to Top