APPENDIX

Pillow textures apparently affected core recovery at Site 1243. The pillows within the igneous sequences of Hole 1243B show the following zonal characteristics from their upper outer crust to their interiors:

1. A black to dark gray glass zone commonly <5 mm thick;
2. A dark gray to brownish gray variolitic zone normally 5-20 mm thick;
3. A gray to pale brown vesicular zone ~20-40 mm thick,
4. A brown to dark brown transitional zone ordinarily up to 100 mm thick; and
5. A dark green crystalline interior.

Lower outer crusts are similarly texturally zoned, but the thickness of the each zone is approximately half or less than those of the upper part. Glass zones are good indicators of pillow boundaries. They may be present on the tops of the pillows (upper glass zone [UGZ]), on the bottoms (lower glass zone [LGZ]), or on both (upper and lower glass zone [ULGZ]), depending upon the mode of pillow formation.

Cored samples were assigned piece numbers. Hole 1243B yielded a total of 359 pieces (with rock samples numbered 5A, 5B, and 5C are counted as one piece) (Table AT1). A glass zone is observed in 46 pieces of rock (12.8%). Among those 46 samples, only 11 pieces can be clearly oriented and assigned to the appropriate lithologic unit; nine pieces (19.6%) are from the UGZ, and two pieces (4.3%) are from the LGZ. No rock piece from the ULGZ was recovered from this hole. Interpillow fragments, including pillow breccia as well as hyaloclastite, were not recovered from Hole 1243B, except for one piece <10 mm in diameter of pillow glass cemented with clay materials (from the cuttings in Section 203-1243B-19B-1). The observed ratio of UGZ to LGZ is 5/1. Two reasons are proposed for the bias toward more glassy tops than bottoms—pillow-lava formation or drilling mechanics or both.

1. Pillow-lava formation: when lava erupts onto the seafloor, the UGZ forms by chilling the hot lava as it comes into contact with cold seawater. On the other hand, the chilling effect on the pillow bottom is less abrupt and hence the LGZ is thinner, because either a pre-existing underlying pillow was still hot or contact with seawater was too brief to chill the lava of the pillow bottom rapidly.
2. Drilling mechanics: abyssal pillows contain many visible radial joints induced by the cooling of magma from the pillow surface to the interior. In addition, incipient radial joints may be induced by the stress of cooling. The rotary core barrel (RCB) bit is 25 cm in diameter, but its throat is only 6 cm in diameter. During drilling with the RCB, the maximum recovered core volume is only ~6% of the volume of rock penetrated by the bit. The proportion of cored rock to cuttings is very small, and any fractures increase the tendency of the rock to be crushed into cuttings rather than to be transported into the core barrel. This may be one of the primary reasons it is difficult to recover full cores with the RCB, except under very good rock conditions such as coring a massive joint-free rock with moderate hardness.

Pillows form with so many joints that pillow lavas break very easily into drill core pieces at these joints before any drilled cores are pushed into the core barrel. Coring may be more effective in the lower part than in the upper part of the pillow because the glass zone at pillow-on-pillow contacts is far weaker and smoother than is the pillow interior. This may be one of main reasons why we recognize more pillow pieces with UGZ than those with LGZ. Generally speaking, the recovery rates of basement rocks in Hole 1243B are relatively poor, as the rocks collected total only 22.55 m in 85.3 m of penetration. Recovery in three cores (Cores 203-1243B-4R, 7R, and 9R) is 48%-64%. Recovery in the other 14 basement cores reached a maximum of 35% and averaged 20%.

As shown in Table AT1, core recovery rates in Hole 1243B correlate well with the degree of alteration and the existence of visible secondary veins. The three cores mentioned as having good recovery are moderately altered and show fillings. Both alteration and secondary precipitation in some veins were induced by ocean floor diagenesis under low-temperature conditions. Hydrothermal alteration was not observed among the drilled rocks. On the other hand, the rock specimens of the 14 other cores are fresh or only slightly altered and free from joint filling by secondary precipitation. The alteration may weaken rock hardness, and secondary joint filling makes massive and solid rocks out of fragile highly jointed ones. Both effects contribute to the increase in core recovery rates.

³Ocean Research Institute, University of Tokyo, 1-15-1 Minamidai, Nakano-ku, Tokyo 164-8639 Japan. ishii@ori.u-tokyo.ac.jp