We have carried out 21 cm radio continuum observations of 16 ultracompact H II regions using the VLA (DnC array) in a search for associated extended emission. We have also observed the H76 alpha recombination line toward all the sources and the He76 alpha line at the positions with strong H76 alpha line emission. The UC H II regions have simple morphologies and large (greater than or similar to 10) ratios of single-dish to VLA fluxes. We detected extended emission toward all the sources. The extended emission consists of one to several compact (similar to1' or 0.5-5 pc) components and a diffuse extended (2'-12' or 4-19 pc) envelope. All the UC H II regions but two are located in the compact components, where the UC H II regions always correspond to their peaks. The compact components with UC H II regions are usually smaller and denser than those without UC H II regions. For individual sources, we derive the spectral types (O7-O4) of the ionizing stars and the fractions of UV photons absorbed by dust within the nebulae, which are significantly different from previous estimates based on the UC H II regions alone. Our recombination line observations show that the ultracompact, compact, and extended components have approximately the same velocity in the individual sources with one exception (G25.72+0.05), implying that they are physically associated. The compact components in each object appear to be ionized by separate sources, while the UC H II regions and their associated compact components are likely to be ionized by the same sources on the basis of the morphological relations mentioned above. This suggests that almost all of the observed UC H II regions are not "real" UC H II regions but ultracompact cores of more extended H II regions, and that their actual ages are much greater than their dynamical age (less than or similar to 10(4) yr). We find that most of simple UC H II regions previously known have large ratios of single-dish to VLA fluxes, similar to our sources. Therefore, the "age problem" of UC H II regions does not seem to be as serious as earlier studies argued. We present a simple model in which the coexistence of the ultracompact, compact, and extended components for a long (>10(5) yr) time is easily explained by combining the champagne flow model with the hierarchical structure of massive star-forming regions. The well-known relation between the density and diameter of H II regions, n(e) proportional to D-1, is a natural consequence of the hierarchical structure according to our model. We discuss some individual sources.
