Length-frequency data are widely used to derive growth estimates, especially in small tropical fishes (see the POPGROWTH table, this vol.). Froese and Binohlan (2000) have shown that length-frequency curves can also be used to get a first assessment of the status of a stock (see Fig. 33) if the data are plotted in a framework of asymptotic length, length at optimum yield, and length at first maturity (see Key Facts, this vol.). With this new table, we try to collect and preserve historical data from unfished or still lightly fished populations, to be contrasted with the curves typically produced from overexploited stocks, where the large, highly fecund fish (the ‘Mega-spawners’) have disappeared and the bulk of the catch is made up of juveniles which had no chance to reproduce.

Over 300 publications with length-frequency distributions were used to extract data for the LENGTH-FREQUENCY table. About a quarter of the available LF data were extracted from growth ‘atlases’, e.g., Ingles and Pauly (1984); Lavapie-Gonzales et al. (1997); Moreau et al. (1995); Dwiponggo et al. (1986); Anon. (1988a); Uosaki and Bayliff (1999). Another quarter are cruise reports, surveys and published chronicles of raw data, e.g., Anon. (1976, 1983, 1984, 1988a, 1988b); Dalzell (1983); Godo and Nedreaas (1986); and Rijavec (1980).

In November 2000, the table contained over 1,500 length-frequency studies with more than 9,600 length-frequencies in the linked LFDATA table. This covers more than 500 species in 136 families, 150 of which are freshwater, 184 brackish water and 441 saltwater species from 95 countries. Over one-third of the species so far covered belong to unexploited stocks. The coverage (i.e., in number of species) of related parameters pertinent to fisheries management in the current dataset is as follows: asymptotic length, L¥ (42%); constants of the length-weight relationship, a and b (32%); length at which yield is at its optimum, L_opt (30%); total mortality, Z (29%); natural mortality, M (26%); length at first capture, L_c (14%); fishing mortality, F (8%); length at first maturity, L_m (5%).

Main Ref.: Numeric field referring to the published source of frequency distribution data. Double-clicking on this field opens the REFERENCES INFORMATION window, which gives the bibliographic details, i.e., author, year, title and source.

Locality/Country: Text field specifying the locality and country of the sampling station or area. Double-clicking on the Country field opens the COUNTRY REFERENCE window, which gives further details on the country in question.

Latitude, Longitude and Accuracy: Numeric fields referring to the coordinates of sampling station or area obtained either from a geographic positioning system (GPS) or specified from a map and include their level of accuracy (see OCCURRENCES table, this volume).

Depth and Temperature: Numeric fields specifying the depth range (m) and temperature (ºC) of the water column sampled.

Gear: Choice field indicating the type of gear used and consists of the following options: seines; trawls; dredges; liftnets; castnets; gillnets; traps; hooks and lines; various gears; other.

Sex: Choice field indicating the sex of the fish sampled and consists of the following options: females; males; unsexed/mixed fish.

Lm: Numeric field, which gives the value of length at first maturity obtained by or cited in the study.

L¥ : Numeric field, which gives the value of the asymptotic length of the fish population. Further categorization is provided to indicate if this value is a calculated result from the study or a value obtained from the Key Facts page of FishBase.

Length type: Length of fish measured, e.g., total length.

Length range: Numeric fields specifying the range of lengths (cm), from smallest fish to largest fish sampled, of all frequency distribution samples in the study.

Length/weight parameters: Numeric fields a and b indicating the constants of the length-weight relationship obtained or calculated directly from the samples.

Frequency type: Choice field describing the type of the frequency distribution and includes four options: absolute number measured (i.e., raw data); % of sample (i.e., frequency expressed as a fraction of the total number of fishes sampled); raised to the catch (i.e., frequency weighted by the total catch); other.

Year: Numeric fields indicating the period of sampling.

Comments: Text field providing further descriptions on locality or sampling conditions, gear type (if the ‘other’ option is ticked), frequency type (if the ‘other’ option is ticked), and other pertinent information.

LF code: Numeric field used internally to link each sample to the specific study.

Date of sampling: Numeric field indicating the exact date of sampling with the format dd/mm/yy. Note that in cases where the exact day of sampling is not available or where several samples (or sampling days) were lumped into a month, date of sampling is fixed at the 15^th of the month in question.

L/F data: Numeric fields indicating the mid-length and the number of fish sampled in that length class.

You get to view length-frequency data by clicking on the Population dynamics button in the BIOLOGY window and the Length-frequency button in the POPULATION DYNAMICS window. Double-click on any row in the LIST OF FREQUENCY STUDIES window, then on the LF data button in the LENGTH-FREQUENCY window. The graph button in the LENGTH-FREQUENCY window displays a frequency distribution graph of all samples for a particular locality or study. On the other hand, the Graph button in FREQUENCY DISTRIBUTION window displays the frequency distribution graph for a single sample in the study.

The LENGTH-FREQUENCY table was not yet available in the Internet in November 2000.