Martini et at: A population profile for Myxine glutinosa 
521 
Table 3 
Results of post-hoc multiple comparison tests (Scheffe F-test) at the 5% level of significance. For counts, means, and standard 
deviations, refer to Table 2. The asterisks (*) indicate values significant at the 5% level (Scheffe F-test). 
Groups compared 
Mean diff. 
P-value 
Groups compared 
Mean diff. 
P-value 
Prebranchial slime pore counts, as value and percentage of 
IGM 
vs. OGM 
-0.276 
(-0.010) 
0.9987 
(0.0694) 
total (in parentheses) 
MAC 
vs. NWA 
0.660 
(0.001) 
0.9538 
(0.9995) 
IGM vs. MAC 
5.216 
(0.020) 
<0.0001* 
(0.4127) 
MAC 
vs. OGM 
-1.271 
(-0.004) 
0.2460 
(0.9376) 
IGM vs. NWA 
7.517 
(0.033) 
<0.0001* 
(0.1905) 
MAC 
vs. ENA 
-0.201 
(-0.006) 
0.9940 
(0.1040) 
IGM vs. OGM 
2.756 
(0.005) 
0.2380 
(0.9979) 
NWA 
vs. ENA 
-0.861 
(-0.007) 
0.8116 
(0.6125) 
IGM vs. ENA 
6.045 
(0.014) 
<0.0001* 
(0.0915) 
NWA 
vs. OGM 
-1.932 
(-0.006) 
0.1156 
(0.9204) 
MAC vs. NWA 
2.301 
(0.019) 
0.7105 
(0.8340) 
OGM 
vs. ENA 
1.075 
(-0.001) 
0.3745 
(>0.9999) 
MAC vs. OGM 
-2.46 
(0.009) 
0.4471 
(0.9913) 
MAC vs. ENA 
.828 
(-0.010) 
0.8798 
00.9999) 
Total slime pore counts 
NWA vs. ENA 
-1.472 
(-0.019) 
0.9448 
(0.8091) 
IGM 
vs. MAC 
14.227 
<0.0001* 
OGM vs. NWA 
4.761 
(0.028) 
0.0865 
(0.6534) 
IGM 
vs. NWA 
18.429 
<0.0001* 
OGM vs. ENA 
3.288 
(0.009) 
0.0661 
(0.9849) 
IGM 
vs. OGM 
7.566 
0.0039* 
IGM 
vs. ENA 
17.38 
<0.0001* 
Trunk slime pore counts as 
value and 
percentage 
of total 
MAC 
vs. NWA 
4.203 
0.6730 
(in parentheses) 
MAC 
vs. OGM 
-6.661 
0.0353* 
IGM vs. MAC 
8.550 
(0.002) 
<0.0001* 
(0.9985) 
MAC 
vs. ENA 
3.154 
0.0922 
IGM vs. NWA 
9.197 
(-0.017) 
<0.0001* 
(0.2559) 
NWA 
vs. OGM 
-10.863 
0.0054* 
IGM vs. OGM 
4.838 
(0.004) 
0.0022* 
(0.9985) 
NWA 
vs. ENA 
-1.076 
0.9995 
IGM vs. ENA 
11.170 
(0.010) 
<0.0001* 
(0.0212)* 
OGM 
vs. ENA 
9.787 
<0.0001* 
MAC vs. NWA 
0.647 
(-0.019) 
0.9997 
(0.2076) 
MAC vs. OGM 
-3.712 
(0.002) 
0.0861 
(>0.9999) 
Total cusp counts 
MAC vs. ENA 
2.619 
(0.008) 
0.0030* 
(0.3125) 
IGM 
vs. MAC 
0.432 
0.9625 
NWA vs. ENA 
1.972 
(0.027) 
0.8591 
(0.0054)* 
IGM 
vs. NWA 
3.409 
0.0006* 
NWA vs. OGM 
-4.359 
(0.022) 
0.2478 
(0.2986) 
IGM 
vs. OGM 
0.235 
>0.9999 
OGM vs. ENA 
6.331 
(0.006) 
<0.0001* 
(0.9760) 
IGM 
vs. ENA 
1.883 
<0.0001* 
MAC 
vs. NWA 
2.978 
0.0109* 
Tail slime pore counts as value and percentage of total 
MAC 
vs. ENA 
1.667 
0.0003* 
(in parentheses) 
MAC 
vs. OGM 
-0.197 
>0.9999 
IGM vs. MAC 
0.995 
(-0.006) 
0.0166* 
(0.0694) 
NWA 
vs. ENA 
-1.310 
0.7119 
IGM vs. NWA 
1.655 
(-0.005) 
0.1012 
(0.9683) 
NWA 
vs. OGM 
-3.175 
0.1263 
IGM vs. ENA 
0.794 
(-0.012) 
0.0088* 
(<.0001)* 
OGM 
vs. ENA 
1.864 
0.4313 
between the groups IGM-MAC, IGM-NWA, IGM- 
OGM, and MAC-ENA were not significant when com- 
pared as percentages of the total slime pore count. 
Total cusp counts differed significantly between 
NWA-IGM, NWA-MAC, IGM-ENA, and MAC-ENA. 
Discussion 
The primary goal of this analysis was to assess the 
validity of the proposed splitting of M. glutinosa L. into 
two species. With the exception of the IGM data, the 
proposal by Wisner and McMillan ( 1995) was based on 
a general morphological comparison of animals col- 
lected from the eastern and western North Atlantic. 
This study used their data, assigned to the NWA, OGM, 
MAC, and ENA groups. As is often the case when work- 
ing with fishes whose lifestyles, habits, and population 
dynamics are poorly understood, there are a number 
of potential complicating factors that could bias these 
data. For example, both the Wisner and McMillan study 
and our own have compared specimens collected by 1) 
different methods, 2) at different times, and 3) at dif- 
ferent depths. This is not unusual; the majority of the 
59 currently recognized species of hagfishes are known 
from small numbers of animals (often just one) caught 
accidentally in mobile fisheries gear. The limitations 
of our data sets are therefore shared not just with 
Wisner and McMillan ( 1995 ) but with many other stud- 
ies of hagfish systematics. Our goal was to determine 
if — given the limitations of the available data — the pro- 
posed split could be justified on the basis of the avail- 
able morphological data. 
We will now discuss each of these complicating 
factors as they influence hagfish collections in gen- 
eral, and the data in this study in particular. 
The collection method might affect 
the size range of animals captured 
Comparative data on trawl versus trap collection of 
Myxine are unavailable, but it is known that trawl- 
