Scanning electron micrographs were obtained on Ni samples irradiated for 0, 15, and 120 min, as shown in Figure 2. As the malleable particles are irradiated, profound changes in particle aggregation and morphology are observed. The surface of our Ni powder is initially highly crystalline, but upon sonication the surface is smoothed quite rapidly. At the same time, the extent of aggregation increases dramatically. We believe that both effects are due to interparticle collisions driven by the turbulent flow created by the ultrasonic field. The increase aggregation accounts for the eventual decrease in the observed surface area and probably also causes the small diminution in activity observed after lengthy sonication. This change in surface morphology is associated with a dramatic change in surface composition. Initially, a thick oxide coat is found (with a surface Ni/O ratio of 1.0) extending ~250 Å into the particle. After 1 h of ultrasonic irradiation in octane, the oxide layer is much thinner (60 Å, with a surface Ni/O ratio of 2.0). In fact, most of the oxide layer in the irradiated sample is due to its air exposure during sample transfer; the oxide coating is fully reestablished after ~15 min of air exposure. It is likely that the origin of our observed sonocatalytic activity comes from the removal (through interparticle collisions) of the surface oxide layer normally found on Ni powders. A clean Ni surface is an active catalyst; nickel powder with its usual surface oxide coating is not. Recently we have described the isolation and partial structure elucidation of esperamicins A₁ and A₂ which are produced by cultures of Actinomadura verrucosospora (BBM 1675, ATCC 39334) and are characterized by broad spectrum antitumor activity in murine systems. To our knowledge they are the most potent antitumor agents yet discovered. The esperimicins are produced as a complex of related compounds which we have resolved into a number of components, A₁, A₁b, A₂, A₄, B₁, and B₂. In addition to these bioactive metabolites, we have also isolated esperamicin X (1), an inactive compound coproduced by the organism. Similarities in the physical properties of 1 with the bioactive metabolites have led us to undertake the structure elucidation of 1 (Figure 1). Compound 1 was isolated as a white crystalline solid, mp 182-184 °C, [α]D -36° (c 0.5, CHCl₃). Its molecular formula was established as C₃₁H₃₂N₂O₁₄S by using FAB mass spectroscopy (MW 756) and elemental analysis. The IR spectrum of 1 had bands characteristic of hydroxyl, ester, amide, and enol ether functions. The UV spectrum of 1 was similar to that of esperamicin A₁. Examination of the ¹H and ¹³C NMR spectra of 1 showed the presence of numerous resonances observed in the spectrum of esperamicins A₁, A₁b, and A₂ and encouraged a more thorough analysis. Methanolysis of 1 yielded the 2-deoxy-L-fucose derivative 2 identical in all respects with that obtained on methanolysis of esperamicins A₁ and A₂ as well as a new product, 3. Compound 3 was obtained as white crystals, mp 223 °C. The IR spectrum of 3 showed bands for ketone, urethane, and hydroxyl functions. The UV spectrum of 3 was uninformative with only very weak absorbances above 230 nm. A molecular formula of C₁₆H₁₉NO₆S (MW 363) was determined by high-resolution mass spectroscopy. From the ¹H and ¹³C NMR spectra, the presence of a number of substructural fragments could be deduced, i.e., a disubstituted aromatic ring, an allylic group =CHCH₂S, three isolated methine groups bearing heteroatoms, a ketone carbonyl, and two heteroatom substituted quaternary carbons. Assignment of an unambiguous structure based on the data was not possible; consequently, crystals of 3 grown from methanol-chloroform were subjected to X-ray analysis. The X-ray experiment defined only the relative, not the absolute, stereochemistry. The tetracyclic core of the molecule can be dissected into smaller rings to discuss the conformation. It remained to establish the points of attachment of the 2-deoxy-L-fucose fragment to the core. The mass spectra of compounds 1 and 3 permit us to assign the point of attachment as being at C4. Further support for this assignment from ¹H and ¹³C NMR comparisons of 1 and 3 was available. The assignment of the α-glycosidic linkage in 1 was made on the basis of the C1'-H coupling constants to the C2' protons. With the structure of esperamicin X (1) in hand, assignment of the NMR spectra data to specific structural features was accomplished. Comparison of the spectra of esperamicin X with those of esperamicin A₁ revealed numerous similarities between them. Reconciliation of these structural similarities and differences between 1 and esperamicin A₁ is the subject of the following communication in this issue. In the preceding communication in this issue, the structure elucidation of esperamicin X was described. We now report the structure elucidations of esperamicins A₁, A₂, and A₁b (compounds 1a-c, respectively, Figure 1) through chemical degradation and the analysis of the spectra of the degradation products. Esperamicin A₁ (1a) contains four sugars and an aromatic chromophore which are attached at two points to a bicyclic core. Of the four sugars in 1a, three have not previously been reported. The central core contains a number of unique functionalities within a bicyclo[3.7.1] system; an allylic trisulfide attached to the bridging atom, a 1,5-diyn-3-ene system, and an α,β-unsaturated ketone.