Hinton et at: Use of gonad indices to estimate the reproductive status of Xiphias gladius 
it is necessary to ensure that the averages being com- 
pared are for individuals with comparable reproduc- 
tive potential or maturity. Thus, it is standard prac- 
tice to use a minimum length (e.g. Miyabe and Bayliff, 
1987; Sosa-Nishizaki, 1990; Nakano and Bayliff, 
1992, Arocha et al., 1994; Arocha and Lee, 1995) to 
decide which data to include in estimates of average 
values of GI, making it important to document that 
minimum-length criteria have no impact on meth- 
ods used to estimate reproductive status (Cayre and 
Laloe, 1986). 
Data and methods 
Details on data collection and histological analyses, 
other than estimation of the values of individual go- 
nad indices, may be found in Taylor and Murphy 
(1992). Female swordfish were assigned to eight de- 
velopmental classes (Murphy and Taylor, 1990) based 
on the appearance of histological features (Wallace 
and Selman, 1981). These classes and mean observed 
oocyte diameters were 1) immature, < 20 pmm; 2) 
developing, 71 pmm; 3) maturing, 160 pmm; 4) ma- 
ture, 434 pmm, 5) gravid, 723 pmm; 6) spawning or 
partially spent, 823 pmm; and 7) spent, 181 pmm. 
Individuals in class 8 (recovering) were observed but 
not described in Taylor and Murphy (1992). Gonads 
of swordfish in class 8 exhibited signs of having 
spawned in the previous season and were undergo- 
ing maturational, prespawning development for sub- 
sequent reproductive efforts. 
The preferred formulation for GI may be deter- 
mined by examining the relation of gonad weight to 
measures of body size (de Vlaming et al., 1982). The 
formulation chosen should meet the underlying as- 
sumptions (de Vlaming et al., 1982) for use of GI as 
an index of reproductive status. In addition to exam- 
ining the previously described “standard” expression 
of GI (hereafter referred to as GI(1)), we examined 
GI - \n(GW)I\n(EFL), hereafter referred to as GI(2), 
and GI = GW/EFL. Stepwise analysis of covariance 
(ANCOVA) was used to examine how well these for- 
mulations for GI met the underlying assumptions 
(de Vlaming et al., 1982) for use of GI as an index of 
the reproductive status of female swordfish. Values 
of GI were determined for the fish for which histo- 
logical data had been obtained. For those individu- 
als for which measurements of EFL were not ob- 
tained, measurements of LJFL were used to estimate 
EFL as follows: EFL = -8.259 + 0.930 x LJFL [n= 316, 
r 2 =0.996,P<0.001] (Taylor and Murphy, 1992). Of the 
over 400 fish examined by Taylor and Murphy ( 1992), 
there were 85 individuals (Table 1) with measure- 
ments (40) or estimates (45) of EFL ranging from 73 
to 253 cm, for which there were GW’s and data from 
Table 1 
The status of reproductive activity (R) of swordfish determined by histological analyses [Taylor and Murphy, 1992]), EFL = eye 
fork length (cm), and GW = gonad weight (gm). 
R 
GW 
EFL 
R 
GW 
EFL 
R 
GW 
EFL 
R 
GW 
EFL 
2 
3 
77 
2 
100 
113 
3 
752 
169 
6 
8,740 
221 
2 
13 
86 
2 
110 
147 
4 
530 
161 
6 
8,840 
208 
2 
14 
88 
2 
135 
115 
4 
780 
182 
6 
9,920 
206 
2 
15 
94 
2 
140 
118 
4 
800 
187 
6 
10,180 
188 
2 
30 
96 
2 
140 
128 
4 
1,320 
186 
6 
10,430 
227 
2 
30 
95 
2 
150 
121 
4 
2,140 
219 
6 
11,340 
223 
2 
30 
92 
2 
220 
93 
4 
2,888 
201 
6 
15,140 
253 
2 
35 
97 
2 
225 
73 
5 
2,690 
169 
8 
446 
167 
2 
35 
99 
2 
255 
148 
5 
3,950 
184 
8 
540 
172 
2 
35 
106 
3 
100 
121 
6 
1,240 
181 
8 
600 
166 
2 
35 
106 
3 
105 
130 
6 
1,540 
202 
8 
640 
164 
2 
37 
101 
3 
no 
123 
6 
1,760 
174 
8 
730 
183 
2 
45 
114 
3 
200 
137 
6 
2,270 
182 
8 
800 
190 
2 
50 
99 
3 
240 
139 
6 
3,650 
181 
8 
980 
179 
2 
50 
96 
3 
300 
168 
6 
3,780 
171 
8 
995 
197 
2 
55 
104 
3 
353 
166 
6 
3,850 
208 
8 
1,325 
200 
2 
70 
103 
3 
437 
120 
6 
3,900 
155 
8 
1,340 
222 
2 
70 
105 
3 
470 
184 
6 
4,000 
180 
8 
1,360 
249 
2 
80 
104 
3 
500 
181 
6 
4,700 
191 
8 
1,500 
211 
2 
80 
112 
3 
620 
185 
6 
4,720 
197 
8 
1,790 
191 
2 
90 
113 
3 
680 
186 
6 
6,034 
154 
2 
100 
128 
3 
680 
174 
