445 
theless be safely disposed at sea, as in the case of substances permitted 
to be dumped under Annex II. The focus of this consideration is on "“tech- 
niques" in dumping, rather than upon the “intrinsic properties" of the ma- 
terial to be dumped.*/ 
This use of “special care” is the basis for classifying substances to An- 
nexes I or II. The “General Guidelines for Classification of Substances to 
Annexes I and II to the London Dumping Convention"-- which were approved by 
Contracting Parties at the Fifth Consultative Meeting. LDV V/WP. 5/Add. 1, 
para. 3.2.1, Annex--classify substances to Annex I if they are "... simul- 
taneously toxic, persistent and bioaccumulative." Annex II substances are 
similarly defined as "... those which exhibit one or more of the properties 
of toxicity, persistence of bioaccumulation, but which may be safely dis- 
posed of in the marine environment if special care is used in the dispos- 
al." (Emphasis added) Under these classification criteria, an Annex II 
substance can have all of the Annex I properties of toxicity, persistence, 
and bioaccumulation. It is nevertheless classified to Annex II, rather 
than to Annex I, because it may be "safely disposed in the marine environ- 
ment if special care is used in the disposal." The focus of the classifi- 
cation is upon “special care” and “safe disposal“, and not upon the proper- 
ties of toxicity, persistence, and bioaccumulation, per se. 
SCIENTIFIC BASIS FOR APPLICATION OF CRITERIA 
Although the classification of substances to Annexes I or II has been on a 
categorical basis, scientifically the criteria can equally be applied on a 
*/ Under the Interim Guidelines certain of the "special care" measures 
considered in this submisston may also be relevant in determining 
"harmlessness" and "trace contaminants". Paragraph 5 of Appendix I 
provides that "in applying the results of tests to predict the envir- 
onmental impact of the proposed disposal, the method of disposal and 
the dilution of the waste that would result after dwnping should be 
considered..." 
